subslover
3 months ago
MicroAlgo Inc. (NASDAQ: MLGO) Announced to Jointly Establish a Micro-Consciousness Quantum Research Center With WIMI (NASDAQ: WIMI)
BEIJING, June 4, 2024 /PRNewswire/ -- MicroAlgo Inc. (NASDAQ: MLGO) (the "Company" or "MicroAlgo"), today announced that MicroAlgo and WiMi will jointly establish a micro-consciousness quantum research center. It will integrate physics, mathematics, medicine, genetics, computer science, biology, polymer chemistry, philosophy, psychology, sociology, and many other disciplines to form a comprehensive research institute based on various disciplines centered on consciousness.
Research in basic science is a long process that requires a long-term commitment from scientists. At the same time, it also requires the support and promotion from the society and enterprises. Against the backdrop, MicroAlgo (NASDAQ: MLGO) and WiMi (NASDAQ: WIMI) plan to establish a micro-consciousness quantum research center together, aiming to utilize the characteristics of quantum theory, combined with the most cutting-edge artificial intelligence technology, to study the relationship between consciousness and quantum. Continuously explore the deep science, to unify the quantum and human consciousness, and expand the limitations of science.
The micro-consciousness quantum research center builds a bridge between basic science and applied technology, closely connecting the theoretical research of basic science and the practice of applied technology. It not only focuses on basic science research, but also promotes a arrange of innovative technologies in practice, and actively facilitates the transfer from research results to practical applications. From the vision of globalization, the center will integrate the world's top resources and gather talents from the world, and is committed to becoming an international innovation platform for quantum information science, artificial intelligence, neuroscience and biology, and to lead towards a future where artificial intelligence and quantum science will go hand in hand. The core research directions of the micro-consciousness quantum research center are as follows.
1. Quantum computing and consciousness model: Utilizing the principles and advantages of quantum computing to research the nature of the conscious phenomenon. Build a quantum algorithm to simulate the dynamic behavior of neuronal networks and the brain's activities in cognition, perception, and decision-making. The goal is to develop quantum models of consciousness capable of performing highly complex tasks, which may include, but are not limited to, emotion recognition, creative thinking simulations, and quantitative analysis of states of consciousness.
2. Brain-computer interface technology: Exploring the theoretical framework and practice for the integration of consciousness and machine, researching and developing the next-generation intelligent interface. Develop high-precision quantum sensors for monitoring brain activities. At the same time, utilizing quantum encryption technology to ensure the security of data transmission during brain-computer interaction. In addition, explores the use of quantum computing processing capabilities to decode brain signals in real-time to achieve more efficient human-machine interaction, laying a solid foundation for future human-machine applications.
3. Big data AI and quantum consciousness: Simulating the working mechanism of the human brain, combining quantum computing with artificial intelligence to build a new generation of intelligent systems, creating a quantum-enhanced AI system, and unlocking the mysteries of consciousness and the quantum world. Promote the application of quantum computing in simulating brain functions and optimizing machine learning algorithms, and explore the use of quantum features to enhance the cognitive ability and creativity of AI, in a bid to develop a humanoid AI system based on quantum consciousness to open a new era of intelligence.
4. Quantum-driven generative consciousness: Explore the potential connection between the conscious phenomenon and the principles of quantum physics, understand the nature of consciousness at the quantum level, construct a cross-disciplinary technical theoretical system, and provide theoretical support for the development of consciousness science.
Micro-consciousness quantum research center establishes a deep alliance with top universities and research institutes in China based on mutual trust, and the modes of cooperation include joint research and development, building joint laboratories, building industrial institutes, building engineering experiment centers, and training talents together. Specifically, through the joint R&D project, scientific research resources can be shared, and the resource advantages of both sides can be pooled together to overcome the major scientific problems in the field of consciousness quantum science. Through the construction of joint laboratories and industrial institutes, theoretical research and practical teaching can be closely integrated to provide researchers with learning and research opportunities close to the frontiers of the industry. At the same time, accelerates the transfer from scientific research results to industrial applications. The joint construction of the engineering experiment center provides a platform for the development and testing of new technologies and accelerates the development of technology. The micro-consciousness quantum research center will bridge the gap between theoretical research and practical applications, and promote human society to step into a new era that is smarter, more virtual and more efficient.
About MicroAlgo Inc.
tw0122
7 months ago
Hey at least WIMI has more substantial news then HOLO.
HOLO joins an organization and pumps 2000 percent hilarious but good money maker. Recent News here you would think this could be a 5000 percent winner lol..
Beijing, Feb. 05, 2024 (GLOBE NEWSWIRE) -- WiMi Hologram Cloud Inc. (NASDAQ: WIMI) ("WiMi" or the "Company"), a leading global Hologram Augmented Reality ("AR") Technology provider, today announced that a multi-view representation learning algorithm is to deal with the data stream clustering problem. The multi-view representation learning algorithm can provide an effective solution to the data stream clustering problem. The multi-view representation learning algorithm is a method of learning and fusing data from multiple views to obtain a more comprehensive representation. In data stream clustering, multiple views can be used to represent different aspects of the data stream, such as time series view, spatial view, etc., and each view can provide different information.
By learning the features of each view, the potential patterns and structures of the data are discovered and fused to improve the accuracy and stability of data stream clustering for better understanding and analyzing the data stream. Currently, multi-view representation learning algorithms have been widely used and their prospects are very promising. For example, in the financial field, it can be used for customer segmentation and so on. In the medical field, it can be used for disease diagnosis, patient monitoring, etc. In the field of e-commerce, it can be used for user behavior analysis, product recommendation and so on.
The multi-view representation learning algorithm is able to synthesize information from multiple views to provide a more comprehensive description of the data. Different views provide different features and perspectives, and by combining them, a more accurate and comprehensive representation of the data can be obtained. Since the multi-view representation learning algorithm can utilize information from multiple views, it can provide a richer representation of the data. By fusing multiple views, the algorithm can capture more details and correlations in the data, thus improving the data representation. Multi-view representation learning algorithms can effectively improve the clustering performance of data. By synthesizing information from multiple views, the algorithm can reduce the shortcomings of individual views and improve the accuracy and stability of clustering as a whole. The multi-view representation learning algorithm can better handle noise and outliers in the data, making the clustering results more reliable. The multi-view representation learning algorithm can adapt to different types of data. Since different views can contain different types of features, the multi-view representation learning algorithm can flexibly handle situations with different data types. This makes the algorithm more versatile and adaptable when dealing with multiple data.
It can be seen that multi-view representation learning algorithms have the advantages of synthesizing multi-view information, enhancing data representation, improving clustering performance and adapting to different data types. These advantages make multi-view representation learning algorithms have the potential to be widely used in data clustering tasks.
The dataset, including data from multiple views, is first collected. Pre-processing the data, including data cleaning, feature extraction, and data transformation. Then the data is learned using the multi-view representation learning algorithm to obtain multiple-view representations of the data. The learned multiple views are then clustered to obtain multiple clustering results. The multiple clustering results are integrated to get the final clustering results.
The multi-view representation learning algorithm can be categorized into matrix decomposition-based methods, deep learning-based methods, graph-based methods, etc. Matrix decomposition-based methods can represent multiple views of the data as a matrix, and then use matrix decomposition to learn the data. Deep learning-based methods can utilize models such as deep neural networks to learn the data and get a more accurate representation. Graph-based methods can utilize the ideas of graph theory to learn from the data and get a more comprehensive representation.
The multi-view representation learning algorithm can effectively deal with the data stream clustering problem by jointly learning multiple-view representations and combining them with traditional clustering algorithms. Its core idea is to utilize the information provided by different views to capture the intrinsic structure of the data so as to improve the accuracy and stability of clustering.
In the future, with the continuous development of big data and artificial intelligence technology, the multi-view representation learning algorithm will be applied in more fields. Meanwhile, with the continuous optimization and improvement of the algorithm, its accuracy will be further improved.
About WIMI Hologram Cloud
WIMI Hologram Cloud, Inc. (NASDAQ:WIMI) is a holographic cloud comprehensive technical solution provider that focuses on professional areas including holographic AR automotive HUD software, 3D holographic pulse LiDAR, head-mounted light field holographic equipment, holographic semiconductor, holographic cloud software, holographic car navigation and others. Its services and holographic AR technologies include holographic AR automotive application, 3D holographic pulse LiDAR technology, holographic vision semiconductor technology, holographic software development, holographic AR advertising technology, holographic AR entertainment technology, holographic ARSDK payment, interactive holographic communication and other holographic AR technologies.
subslover
9 months ago
WiMi Developed Explainable Artificial Intelligence (XAI)-based fNIRS Neuroimage Classification
BEIJING, Dec. 7, 2023 /PRNewswire/ -- WiMi Hologram Cloud Inc. (NASDAQ: WIMI) ("WiMi" or the "Company"), a leading global Hologram Augmented Reality ("AR") Technology provider, today announced that it developed explainable artificial intelligence (XAI)-based fNIRS neuroimage classification, bringing a breakthrough in the development of BCI technology. By combining the latest AI technology and BCI parsing, this system is expected to bring advances in BCI technology.
WiMi's XAI-based fNIRS neuroimage classification system consists of several key modules that work together to process, analyze, and interpret data for accurate brain activity classification and interpretation. The system architecture is designed to improve classification accuracy and interpretability, and to ensure the accuracy and utility of the system. The system includes a data preprocessing module for filtering, denoising and normalizing the raw fNIRS data to improve the accuracy of subsequent data analysis.
WiMi's XAI-based fNIRSfNIRS neuroimage classification system employs two key classification modules, i.e., a one-dimensional sliding-window-based convolutional neural network (CNN) and a long short-term memory (LSTM) neural network. These two modules are used to classify different types of brain activity patterns respectively, thus improving the applicability and generalization ability of the system. To address the need for interpretation of model outputs, the system introduces an interpretability module, which employs the machine learning interpretability tool SHapley Additive exPlanations (SHAP) for interpreting the outputs of CNN models. By interpreting the model input variables, the system is able to identify the features that contribute the most to the classification of a particular brain activity, helping researchers to gain insight into the association between brain activity patterns and external device control.
Through these methods and techniques, the system is able to efficiently transform fNIRS data into interpretable classification results. The preprocessing of the data, the application of CNN and LSTM models, and the SHAP interpretation module together form the core of the system, enabling it to improve the accuracy of brain activity classification and provide researchers with interpretable results.
WiMi's XAI-based fNIRSfNIRS neuroimage classification system shows good application prospects and potential. In real brain-controlled robots, prosthesis control and virtual reality scenarios, the system's high-precision classification results provide reliable support for device control and offer new possibilities for the application of BCI technology in medical rehabilitation and virtual reality.
The research and application of WiMi's XAI-based fNIRSfNIRS neuroimage classification system brings new insights to the field of brain science. By parsing brain activity patterns through the interpretation module, the system reveals for researchers the association and mechanism of action between functional regions of the brain, and promotes the development of the entire field of brain science. These important results show that the XAI-based fNIRSfNIRS neuroimage classification system not only improves the classification accuracy of brain activities, but also brings new perspectives to the development and application of BCI. It is foreseeable that it will promote the development and popularization of BCI in the future, and bring a revolutionary change to the interaction between humans and machines.
About WIMI Hologram Cloud
WIMI Hologram Cloud, Inc. (NASDAQ:WIMI) is a holographic cloud comprehensive technical solution provider that focuses on professional areas including holographic AR automotive HUD software, 3D holographic pulse LiDAR, head-mounted light field holographic equipment, holographic semiconductor, holographic cloud software, holographic car navigation and others. Its services and holographic AR technologies include holographic AR automotive application, 3D holographic pulse LiDAR technology, holographic vision semiconductor technology, holographic software development, holographic AR advertising technology, holographic AR entertainment technology, holographic ARSDK payment, interactive holographic communication and other holographic AR technologies.
Safe Harbor Statements
This press release contains "forward-looking statements" within the Private Securities Litigation Reform Act of 1995. These forward-looking statements can be identified by terminology such as "will," "expects," "anticipates," "future," "intends," "plans," "believes," "estimates," and similar statements. Statements that are not historical facts, including statements about the Company's beliefs and expectations, are forward-looking statements. Among other things, the business outlook and quotations from management in this press release and the Company's strategic and operational plans contain forward-looking statements. The Company may also make written or oral forward-looking statements in its periodic reports to the US Securities and Exchange Commission ("SEC") on Forms 20-F and 6-K, in its annual report to shareholders, in press releases, and other written materials, and in oral statements made by its officers, directors or employees to third parties. Forward-looking statements involve inherent risks and uncertainties. Several factors could cause actual results to differ materially from those contained in any forward-looking statement, including but not limited to the following: the Company's goals and strategies; the Company's future business development, financial condition, and results of operations; the expected growth of the AR holographic industry; and the Company's expectations regarding demand for and market acceptance of its products and services.
Further information regarding these and other risks is included in the Company's annual report on Form 20-F and the current report on Form 6-K and other documents filed with the SEC. All information provided in this press release is as of the date of this press release. The Company does not undertake any obligation to update any forward-looking statement except as required under applicable laws.
https://c212.net/c/img/favicon.png?sn=CN87879&sd=2023-12-07 View original content:https://www.prnewswire.com/news-releases/wimi-developed-explainable-artificial-intelligence-xai-based-fnirs-neuroimage-classification-302008619.html
SOURCE WiMi Hologram Cloud Inc.
Back to News Headlines
Other Financial Information
Recent News & Disclosure Filings
Recent SEC Filings
StockpickerCAPC
1 year ago
WIMI innovatively develops a humanoid robot control system.
Source
https://cj.sina.com.cn/articles/view/7651844612/1c815e20402001gn7w?from=finance
June 21, 2023
It is understood that, as a world-leading artificial intelligence company, the research team of WIMI (NASDAQ: WIMI) has developed a brain-computer interface (BCI) based on a head-mounted display (HMD) through extensive research in experiments. Controlled humanoid robot interactive control system. This type of interaction controls the robot to interact with the environment and humans through Steady State Visual Evoked Potentials (SSVEP). In this solution, the robot's embedded camera provides real-time feedback, and the stimulus feedback is integrated into the HMD display.
In this research, WIMI's research on the humanoid robot control system based on the head-mounted display (HMD) through the brain-computer interface (BCI) demonstrated a new interaction of the BCI-controlled humanoid robot based on the head-mounted display. It provides a more natural and intuitive way to control the robot. Experimental results show that the control system can provide precise control signals, and users have a very good interaction experience with it. This control method has potential and can be used in many scenarios that require precise control, such as medical care, education, entertainment and other fields.
Humanoid robots have achieved rapid development on the basis of continuous changes in control methods and artificial intelligence technologies, and the conditions for commercialization are increasingly mature. WIMI has increased its investment in this field, which shows that it attaches great importance to and is optimistic about humanoid robots. WIMI's navigation assistance solution for BCI-controlled humanoid robots based on head-mounted displays provides users with a novel interaction method that can improve the robot's operating efficiency and interactive experience. In the future, it can bring a more human-like interactive experience and a wider range of application scenarios, and can interact or work collaboratively with people in a designated area.
epilogue
Humanoid robot research started from the imitation of bipedal walking and expanded to the research and development of artificial intelligence. Since the development of the humanoid robot industry chain, the upstream includes raw materials and core components (with a high proportion of value), and the midstream is system integration and body manufacturing. Downstream is subdivided application scenarios (To B & To C), such as education, logistics and mobile, health care and inspection, etc.
Robots are similar to the smart phone industry chain, and are expected to reproduce the changes in the development stage of smart phones. In addition, under the influence of comprehensive factors such as policy support, global aging, and technological progress, the robot industry has fully met the conditions for entering a period of rapid growth. This may be the reason why many high-tech companies attach importance to and take action to enter the market. In general, in recent years, various humanoid robot products have been launched one after another, with different functions and application scenarios. Coupled with the optimism of WIMI, it will undoubtedly further accelerate the development of this field.
StockpickerCAPC
1 year ago
WIMI (NASDAQ:WIMI) Develops Artificial Intelligence Natural Language Generation (NLG) System to Optimize Content Marketing.
Source
https://cj.sina.com.cn/articles/view/7651844612/1c815e20402001glii?from=finance
June 13, 2023
In recent years, with the popularization of the Internet and the acceleration of digital transformation of enterprises, SEO (Search Engine Optimization) has become a key technology for enterprises to improve brand awareness and expand customer base. SEO needs to rely on high-quality website content, and handcrafted content is undoubtedly time-consuming, laborious and inefficient. In order to solve this problem, WIMI developed "Natural Language Generation SEO" to solve this problem. WIMI uses NLG technology to perform semi-automated search engine optimization (SEO) methods for customers' website landing pages, helps customers draft content to support content marketing, and thus greatly reduces the cost of SEO projects and improves ROI (Return on Investment). With the advancement of natural language generation (NLG) technology, technologies such as digital voice assistants and chatbots are developing rapidly. More and more businesses are adopting natural language generation (NLG) technology to support content marketing and optimize SEO results.
The specific implementation process of this semi-automatic method is as follows:
- First, input the website pages that need SEO into the system, and the system will automatically generate basic content. These will be simple descriptions that tell what the page says in terms of relevance and grammar. SEO experts will then review and revise the generated content to ensure its relevance and accuracy.
- Next, use NLG technology to generate more detailed and unique content for the page. During this process, the system analyzes keywords and target audience, and generates highly readable and high-quality content.
- Finally, an SEO expert double-checks and revises the text to ensure its quality and relevance.
WIMI can quickly create high-quality SEO content by using natural language to generate SEO, while avoiding the high cost and low efficiency of manual SEO content. Apart from this, adopting NLG technology also helps to create more attractive and unique content, which is crucial for improving website ranking and attracting more users.
WIMI uses the most advanced NLG semi-automated method and proves that content writing machines can create unique, human-like SEO content. By comparing with traditional human-written SEO text, it was found that modified machine-generated text was almost indistinguishable from text authored by SEO professionals in many human perception dimensions, and SEO content generated using NLG outperformed professional in search engine rankings content created by professionals, while also dramatically reducing the production costs associated with content marketing and increasing ROI.
The technical implementation process of using natural language generation (NLG) technology to support content marketing and optimize SEO results can be divided into the following steps:
Data collection and preparation:
Collect data related to specific fields, including keywords, industry terms, product descriptions, etc. This data will be used as input for training NLG models and generating SEO content.
NLG model training:
Train an NLG model using machine learning and natural language processing techniques. Commonly used methods include neural network-based models such as recurrent neural networks (RNN) or Transformer models. During training, the model learns language patterns, grammar rules, and contextual understanding.
Content generation and optimization:
Use the trained NLG model to generate SEO content. According to pre-set rules and goals, input relevant information and keywords, and the NLG model will generate texts that meet SEO requirements. Generated text can include titles, descriptions, body text, and more.
SEO Content Evaluation:
The generated SEO content is evaluated to ensure its quality and readability. Grammar, spelling, and keyword usage accuracy checks can be done using natural language processing techniques and SEO tools. In addition, human evaluation can be utilized to measure the readability of content and how well it fits the target audience.
Optimization and modification:
According to the evaluation results, optimize and modify the generated SEO content. It can be adjusted for specific keywords to ensure that the content matches the requirements of the search engine algorithm. This process may require several iterations until the resulting SEO content is of the desired quality and effect.
SEO Content Publishing and Tracking:
Publish optimized SEO content to relevant web pages or articles, and track its performance in search engine rankings. Evaluate the effectiveness of NLG-generated SEO content for SEO by monitoring ranking changes and traffic growth, and make necessary adjustments and improvements.
However, it’s worth noting that despite the remarkable results of natural language generation SEO in improving efficiency and reducing costs, there are still some challenges and limitations. The development of NLG technology requires continuous research and innovation to improve the accuracy of language understanding, the quality of data training, and the ability to protect privacy. In addition, manual editing and optimization of generated content is still necessary to ensure its quality and consistency with the brand image.
With the continuous advancement of technology, WIMI's NLG system will become more intelligent and realistic, capable of generating more accurate and personalized SEO content. However, challenges include the accuracy of language understanding, the quality of data training, and privacy protection. Further research and innovation will be key to advancing the application of NLG technology in the field of content marketing.
With the continuous advancement and development of technology, natural language generation SEO is expected to become an important tool in the field of content marketing. Not only does it improve your business' competitiveness and search engine rankings, it also improves user experience and saves time and resources. Leveraging natural language generation SEO techniques to support content marketing and optimize SEO results will become an essential strategy for businesses looking to succeed in the digital age. WIMI holographic natural language generation SEO will continue to be committed to innovation and optimization to provide customers with more efficient and high-quality content marketing solutions. With the continuous development of NLG technology, it will become an important tool in the field of content marketing, helping enterprises to achieve greater success in the highly competitive market.
StockpickerCAPC
1 year ago
WiMi to Develop A Multimodal Information Fusion Detection Algorithm Based on GANs.
Source
https://finance.yahoo.com/news/wimi-develop-multimodal-information-fusion-120000958.html?guccounter=1
June 12, 2023
WiMi Hologram Cloud Inc. (NASDAQ: WIMI), a leading global Hologram Augmented Reality ("AR") Technology provider, today announced that it is developing a multimodal information fusion detection algorithm based on generative adversarial networks(GANs). The multimodal information fusion detection algorithm is a method to improve detection accuracy and robustness by fusing data from different sensors or modalities using a GAN. It is implemented by training two neural networks, a generator and a discriminator, where the generator is responsible for generating false data samples, and the discriminator is responsible for distinguishing between accurate and inaccurate data. The two networks compete with each other for learning until the generator can produce sufficiently realistic data, and the discriminator cannot differentiate between true and false.
In multimodal information fusion detection, data from different sensors or modalities, such as image, sound, and text, can be fused and processed to obtain more comprehensive and accurate detection results. The generator uses local detail features and global semantic features to extract source image details and semantic information. Perceptual loss is added to the discriminator to make the data distribution of the fused image consistent with the source image, which improves the accuracy of the fused image. The fused features enter the interest pool network for coarse classification, the generated candidate frames are mapped to the feature map, and finally, the fully connected layer completes the target classification and localization.
GANs have inherent advantages in image generation, allowing unsupervised fitting and approximation of accurate data distributions. Using generators and discriminators for adversarial purposes allows fused images to retain richer information, and the end-to-end network structure no longer requires the manual design of fusion rules.
The technical process of the GANs-based multimodal information fusion detection algorithm studied by WiMi includes data preprocessing, GANs model training, model testing, result evaluation, and optimization and improvement. Data from different sensors or modalities, such as image, sound, and text, are fused for fusion processing, improving target detection accuracy and robustness. In addition, the end-to-end trained GANs can enhance the complementarity and redundancy between multimodal information features after fusing them to improve the accuracy of target detection and classification based on fused elements.
The multimodal information fusion detection algorithm treats the whole image fusion process as adversarial between a generator and a discriminator. For each modality, a generator and a discriminator can be trained separately. Then, by combining the generated results of multiple modalities, a more accurate and comprehensive detection result can be obtained.
Multimodal information fusion detection algorithm based on GANs is one of the fast-developing research directions in recent years. Much related research has been applied in different fields, such as intelligent surveillance, speech recognition, medical image analysis, industrial inspection, etc.
In the future, WiMi will further explore how to fuse more sensors and modalities to improve the fusion effect and applicability range. At the same time, WiMi will investigate how to adopt more efficient GAN structures and enhance model performance through more effective training methods. In addition, WiMi also considers combining this technique with deep learning to improve the accuracy and robustness of detection further. In conclusion, the multimodal information fusion detection algorithm based on GANs has many application prospects and is a research direction worthy of attention and in-depth study.
StockpickerCAPC
1 year ago
WIMI Holography (NASDAQ: WIMI) launches high-resolution image automatic registration technology based on feature space objects.
Source
https://cj.sina.com.cn/articles/view/7651844612/1c815e20402001glao?from=finance
June 12, 2023
Automatic registration of high-resolution remote sensing images (HRRSIs) has been a challenging problem due to local deformations caused by different shooting angles and lighting conditions. In order to solve this problem, WIMI has proposed a new method based on feature space object (CSO) extraction and matching. First, the CSO and its localization points on the image are automatically extracted using the Mask R-CNN model. Then, each object and its nearest neighbors are encoded by an encoding method based on object category, relative distance and relative orientation. Next, a code matching algorithm is applied to search for the most similar object pairs. Finally, object pairs are filtered by position matching to construct the final control points for automatic image registration. Experimental results show that the proposed method outperforms traditional optimization methods based on local feature points in terms of registration success rate.
With the continuous development of remote sensing technology, the automatic registration of high-resolution remote sensing images (HRRSIs) has been a serious challenge. Different shooting angles and lighting conditions will cause local deformation of the image, which brings difficulties to data processing and analysis. It is reported that WIMI has launched a high-resolution image automatic registration technology based on feature space objects. Provide reliable information support in areas such as monitoring, urban planning, and agricultural management.
The core of this technological innovation is an automatic registration method based on feature-space objects (CSO). Traditional image registration methods usually rely on grayscale registration, transform domain registration or feature point-based registration, but these methods are very sensitive to grayscale, rotation and deformation, and are computationally intensive, making them unsuitable for automatic registration. WIMI holographic technology adopts a brand-new idea, and achieves more accurate registration results by using the Mask R-CNN model to automatically extract CSO and locate its position.
First, the image is scanned using the Mask R-CNN model, and the CSO and its positioning points on the image are automatically extracted. The accuracy and efficiency of this step is based on the accumulation of research and innovation in the field of computer vision by the WIMI Holographic team for many years. Subsequently, each extracted CSO and its nearest neighbors are encoded according to object category, relative distance and relative orientation. The encoded feature vectors provide the basis for subsequent matching.
In order to find the most similar object pair, WIMI Hologram adopts advanced code matching algorithm. The algorithm determines the degree of matching by calculating the similarity between encoded feature vectors. Object pairs with higher similarities are considered as candidates for registration. Further, the initial object pairs are filtered by position matching algorithm to exclude some false matches and obtain more reliable registration results. Through this step, WIMI holographic technology can accurately capture the spatial position relationship in the image, further improving the accuracy and robustness of registration.
According to the data, WIMI's high-resolution image automatic registration technology based on feature space objects has achieved remarkable results in experiments. By testing and comparing multiple data sets, the results show that this technique is significantly better than the traditional optimization method based on local feature points in terms of registration success rate. [color=green]This breakthrough achievement will enable the remote sensing industry to perform data processing and analysis more accurately and efficiently, providing a more reliable basis for decision-making.[/color]
In addition to remote sensing image processing, this technology also has a wide range of application prospects. In the field of urban planning, automatic registration technology based on characteristic spatial objects can help planners better understand urban changes and development trends, so as to formulate more scientific urban development strategies. In terms of environmental monitoring, this technology can provide accurate image registration results, help scientists monitor and evaluate environmental changes, and provide important data support for environmental protection and resource management. In addition, in areas such as agricultural management and disaster monitoring, this technology can also play an important role, providing accurate data analysis and decision support.
At present, WIMI holographic has introduced the high-resolution image automatic registration technology based on feature space objects to the market, and is exploring application scenarios with industry partners. By integrating this technology with existing remote sensing data processing platforms and software, users will be able to easily achieve high-precision image registration, thereby improving the accuracy and efficiency of data analysis. The introduction of high-resolution image automatic registration technology based on feature space objects marks another important breakthrough of WIMI in the field of image processing. The application of this technology will have a huge impact on the remote sensing industry. In the past, image registration required a lot of time and labor, and the results were not necessarily accurate. However, WIMI holographic technology will make the registration process more automated, efficient and accurate, greatly improving the processing efficiency and quality of remote sensing data.
It can be said that WIMI based on the high-resolution image automatic registration technology of feature space objects has solved the long-standing problems in the field of remote sensing image processing. The application of this technology will provide a more reliable and accurate data basis for scientific research and practice in related fields. In addition, WIMI also plans to carry out further research and development work with industry partners. They will work on further optimizing the efficiency and performance of the algorithm, expanding the scope of application of the technology, and developing more application solutions for different fields. This will provide users with more choices and meet the needs of different industries for image processing and data analysis. It provides strong support for accurate data analysis and intelligent decision-making. The successful application of this technology will bring huge economic and social benefits to environmental protection, urban planning, agricultural management and other fields.
StockpickerCAPC
1 year ago
WiMi Hologram Cloud (WIMI) creates a digital holographic AR-HUD solution,
Source
https://www.newstrail.com/wimi-hologram-cloud%EF%BC%88wimi-creates-a-digital-holographic-ar-hud-solution/
June 7, 2023
After years of development, AR has been applied in many fields, such as the introduction of AR filter by Snapchat (SNAP), which adds a new possibility for social entertainment; Baidu (BIDU) map has created a new model of AR innovation and launched the “walking AR navigation” function.
With the continuous improvement of AR technology, a series of new technologies and new trends have emerged and attracted wide attention. For example, AR has also shown great development potential in the automotive field. The emergence of AR-HUD has taken the car driving experience to a new level.
HUD originated in the aviation field, and GM first applied HUD in the automotive field in 1988. As the name suggests, HUD is the projection of important driving information such as speed and navigation onto the windshield, so that the driver can need to see the vehicle information without bowing his head. Its obvious features can better guarantee driving safety. With the update and iteration of the intelligent cockpit, HUD is expected to develop into the core component of intelligent cars with its excellent interactive features of “people and car, people and environment”.
AR-HUD integrates AR function based on HUD, covering a layer of digital images in the real world seen by driving so that the information projected by HUD is integrated with the real driving environment. With the continuous progress of science and technology, more and more automobile manufacturers begin to pay attention to the human-car interaction experience, making the automobile AR-HUD begin to carry richer and more delicate functions and display tasks.
On-board HUD is divided into three main categories
The onboard HUD has experienced three generations of upgrades, continuously optimized imaging quality, continuous increase of information, and greatly enhanced the sense of technology. At present, W-HUD is the mainstream in the market, and AR-HUD is accelerated for mass production.
The first generation is the C-HUD (Combiner HUD), a combined head-up display system. C-HUD uses translucent resin board as the display medium, which has the advantage of convenient installation, but the imaging area is small and the display information is less. Because C-HUD is installed on the vehicle in the form of accessories, it is easy to cause secondary injury to the driver in case of an accident.
The second generation W-HUD (Windshield HUD) windshield head-up display system, which is currently the most widely used HUD, has achieved mass production. The W-HUD uses optical reflection to project driving information onto the front windshield of the car. W-HUD shows a larger range and further projection distance than C-HUD.
The third generation AR-HUD (Augmented Reality HUD) augmented reality head-up display system is a new head-up display technology. Compared with the traditional W-HUD, AR-HUD has a large projection range and more information, which can better combine the data collected by ADAS for scene fusion. Through the superposition of digital images and real scenes, it can enhance the sense of practicality and technology of HUD.
In 2020, the market began to accelerate the iteration of W-HUD to AR-HUD. Because AR-HUD displays increased information compared to W-HUD. AR-HUD collects the data of the external environment through the front-facing radar and camera equipment, calculates the required image and data information through the AR algorithm, and then reflects the image to the windshield through the optical structure in the AR-HUD, and produces a virtual image superimposed on the real object in front of the windshield.
There is no doubt that, as one of the representative features of the intelligent cockpit, AR-HUD is expected to become the final form of HUD in future vehicles. AR-HUD technology has been attracting the attention of the industry. Some institutions expect that the market size of China’s AR-HUD industry will exceed 7 billion yuan in 2023.
Technology is a solid barrier to WiMi
On the eve of the market explosion of AR-HUD, it is reported that WiMi Hologram Cloud (NASDAQ: WIMI) is working on building AR-HUD-related products. Its light field AR-HUD has multi-faceted optical imaging capabilities, which can achieve a better AR fusion effect and a more natural visual experience. In addition, WiMi Hologram Cloud technology solutions are in rapid progress, with uHD key capabilities, to provide rich application scenarios such as instrument information display, AR navigation, safety-assisted driving, night vision/rain and fog enhancement tips, and audio and video entertainment. Project virtual images in different locations in the real world, so that real objects in different positions can visually integrate with HUD virtual images, and may become a dark horse on the track in the future.
WiMi Hologram Cloud As a leading holographic AR application technology provider, fully aware based on AR-HUD technology is gradually become the focus of auto manufacturers, years committed to through independent innovation, positive research, and development, through AR-HUD transmission to holographic AR image display various digital information, an intuitive way to guide driving, improve the intelligent experience, make AR-HUD intelligent cockpit interactive hub, to improve the safety of driving, to help reduce the risk of accident.
With the further development of AR technology, WiMi Hologram Cloud will focus on the core capability of “visual interaction” in the future, while consolidating the dominant position in the field of intelligent automotive electronics and gradually developing products with technological technology such as AR-HUD. It can be said that this is an effort and attempt in emerging industries such as autonomous driving. It is expected to form a perfect human, car, road, and network ecology, and open up a new entrance for the next generation of augmented reality Internet. In addition, WiMi Hologram Cloud will continue to focus on related fields and technologies, and further participate in the larger holographic AR industry track.
To Sum Up
The HUD market is facing a period of rapid development, and many research institutions have given the same view. On-board HUD technology has become an important trend in the automotive industry. Through HUD technology, drivers can get more driving information while maintaining attention and concentration on the road, thus improving driving safety and comfort. With the development and application of AR technology, it is believed that the AR-HUD system can be said to be a major trend in the development of intelligent vehicles in the future, and its addition also makes us full of expectations for future cars.
At present, some people point out that AR-HUD is only a transitional technology to achieve fully autonomous driving in the future. After all, after the realization of autonomous driving, the convenience of AR-HUD head display is also insignificant for users.
StockpickerCAPC
1 year ago
WIMI (NASDAQ: WIMI) develops DPCEngine to improve the efficiency of strategy evaluation through efficient density peak clustering algorithm.
Source
https://cj.sina.com.cn/articles/view/7651844612/1c815e20402001gkoh?from=finance
June 9, 2023
With the rapid development of technologies such as Industry 4.0, CPS (Cyber-Physical System), blockchain, cloud computing, and big data, and the rapid development of today's network and information security, access control has become the key for enterprises to protect data and resource security. The essential. In the context of ever-increasing security requirements for network and information systems, access control plays an important role in the field of network and information security as well as interdisciplinary topics in the Internet of Things. XACML (Extensible Access Control Markup Language) is a standard widely used in the field of access control. When the defined policy set becomes large and complex, the policy evaluation time will increase significantly, and the traditional policy evaluation methods often face performance bottlenecks. Performance for large-scale systems is a challenge. In order to solve this problem, WIMI (NASDAQ: WIMI) has developed DPCEngine, which is an efficient density peak clustering algorithm for improving the performance of strategy evaluation.
It is reported that WIMI WIMI Hologram Institute proposed a policy set clustering method based on the density peak clustering algorithm, which reduces the complexity of policy evaluation by identifying the cluster structure in the policy set. The architecture and algorithm flow of WIMI DPCEngine include key steps such as data preprocessing, density peak clustering, strategy matching and evaluation.
To evaluate the performance and effectiveness of DPCEngine, a real dataset containing a large-scale complex policy set is used for experiments. This dataset contains policies from different domains, covering a variety of access control scenarios. The data set is divided into training set and test set, where the training set is used to build the model of DPCEngine, and the test set is used to evaluate its performance.
WIMI Hologram R&D staff compared DPCEngine with traditional strategy evaluation methods, including methods based on linear search and tree structure.
Two performance metrics are evaluated:
- policy evaluation time and
- matching accuracy.
Policy evaluation time refers to the time required to evaluate an access request, while matching accuracy refers to the consistency between the matching results of DPCEngine and traditional methods.
DPCEngine has a significant performance advantage in policy evaluation time. Compared with traditional methods, DPCEngine can greatly reduce the policy evaluation time, especially when the policy set is large in scale and high in complexity. This is due to the density-peak-based clustering algorithm adopted by DPCEngine, which is able to cluster policy sets into smaller subsets, thereby reducing the search space for evaluation.
In terms of matching accuracy of WIMI's DPCEngine, the experimental results show that there is a high consistency between the matching results of DPCEngine and traditional methods. This shows that DPCEngine does not sacrifice accuracy while improving policy evaluation performance. In addition, we also conducted scalability experiments to evaluate the performance of DPCEngine under different scale policy sets. The results show that DPCEngine can effectively cope with large-scale policy sets and has good scalability. Its working process is shown in the figure.
According to the data, WIMI's DPCEngine, as a strategy evaluation engine based on the density peak clustering algorithm,
has three main functions:
- preprocessing strategy set,
- cluster strategy set and
- matching strategy.
The combined use of these functions can significantly improve the performance and accuracy of policy evaluation.
Preprocessing strategy set:
Before policy evaluation, DPCEngine prepares data by preprocessing strategy sets, making it more suitable for density peak clustering. The preprocessing process includes steps such as data cleaning, feature extraction and data conversion. Ensure data accuracy and consistency by cleaning data to remove redundant, incomplete or erroneous policy information. Avoid negatively affecting the assessment results. The feature extraction process extracts key features from the policy set, such as user roles, resource types, and operation permissions, for subsequent clustering operations. Data transformation converts the strategy set into a data representation suitable for the density peak clustering algorithm, such as a vector or matrix, for cluster analysis.
Cluster policy set:
DPCEngine uses the density peak clustering algorithm to perform cluster operations on the policy set. The density peak clustering algorithm identifies the cluster structure in the policy set by evaluating the density and distance between the policies. The algorithm determines the peak point based on the density and distance between the policies, and divides the policies between the peak points into different clusters . This enables the clustering of a large and complex set of policies into smaller subsets, each cluster representing a set of policies with similar characteristics and behavioral patterns, reducing the time and complexity of policy evaluation. The result of the cluster strategy set is a group of strategy clusters with similar characteristics and behavior patterns. This cluster strategy set method can reduce the time and computational complexity of strategy evaluation, and improve the performance and efficiency of the system.
Matching strategy:
DPCEngine uses clustering results for strategy matching. When an access request arrives, DPCEngine compares and matches it with the pre-generated policy cluster. By searching for the most similar policy in each cluster, DPCEngine can quickly determine the policy set that matches the access request. This clustering-based matching method can significantly speed up policy matching and provide accurate matching results. In addition, DPCEngine can also combine other access control technologies and rule engines to further optimize the policy matching process to ensure system security and compliance.
Through the comprehensive use of preprocessing policy sets, cluster policy sets and matching policies, DPCEngine can provide enterprises with efficient and accurate policy evaluation.
In addition, WIMI's (NASDAQ: WIMI) DPCEngine's preprocessing policy set, cluster policy set and matching policy functions cooperate with each other to provide enterprises with an efficient, accurate and scalable policy evaluation performance. By utilizing the density peak clustering algorithm and the clustering structure of policy sets, DPCEngine can achieve fast policy matching in the case of large-scale complex policy sets. This cluster-based approach reduces the time and computational complexity of policy evaluation and improves the performance and efficiency of the system.
The three main functions of DPCEngine have broad application prospects in enterprises.
- First, the preprocessing policy set function can help enterprises process and prepare huge policy data to ensure data quality and consistency. This is crucial for cleaning and transforming the data prior to policy evaluation to improve the accuracy of subsequent clustering and matching.
- Second, the Cluster Policy Set feature enables enterprises to divide large and complex policy sets into relatively smaller policy clusters with similar characteristics. This cluster operation reduces the size and complexity of policy evaluation and improves the performance and efficiency of the system. By grouping similar policies together, enterprises can match access requests more quickly and implement fine-grained management and control over policies.
- Finally, the Match Policy feature allows organizations to compare and match access requests against pre-generated policy clusters. This clustering-based matching method can quickly locate the policy set that matches the access request, improving the speed and accuracy of policy matching. At the same time, DPCEngine can be used in combination with other access control technologies and rule engines to further optimize policy matching results and ensure system security and compliance.
Currently, the preprocessing policy set, cluster policy set and matching policy functions of WIMI DPCEngine enable enterprises to efficiently evaluate access control policies. This policy evaluation engine based on the density peak clustering algorithm has broad application prospects in various industries and fields, and can help enterprises build a robust security protection system and cope with growing security challenges. With the continuous development and improvement of technology, DPCEngine will further improve the performance and accuracy of policy evaluation, and provide reliable support for enterprises to ensure the security of data and resources.
StockpickerCAPC
1 year ago
WiMi Hologram Cloud is Developing A Memristor-based Neural Signal Analysis System for Efficient BCI.
Source
https://finance.yahoo.com/news/wimi-hologram-cloud-developing-memristor-120000124.html
June 8, 2023
WiMi Hologram Cloud Inc. (NASDAQ: WIMI), a leading global Hologram Augmented Reality ("AR") Technology provider, today announced that it is developing a memristor-based neural signal analysis system to improve signal processing capability, response time and accuracy. Results show a significant improvement in signal processing capability.
The system employs a new computational paradigm closer to how the human brain works, i.e., a memristor-based neural network system. The memristor-based neural network is characterized by high parallelism and low energy consumption. When processing neural signals, the system uses an end-to-end data processing process. That is, the acquired raw signal is directly converted into the final control signal through pre-processing, feature extraction, and classification recognition steps, thus avoiding the frequent data transmission and computation process in traditional architectures and significantly improving the efficiency and accuracy of the system. The memristor arrays can quickly process a large amount of data because they can hold information between each neuron, thus enabling highly parallel processing. This approach is similar to how neurons in the human brain communicate, allowing for more efficient data processing. In addition, memristor-based systems also have higher energy efficiency in storing and reading data, which can significantly reduce power consumption.
Compared with the traditional von Neumann architecture, the system improves brain-computer interface signal processing and linking capabilities hundreds of times. In addition, the system employs innovative hardware architecture and algorithm optimization to efficiently process neural signals from the human brain and convert them into computer-recognizable signals for seamless human-computer connectivity.
The path and manner of implementation of this technology are as follows.
First, an array of memristors needs to be designed and prepared. A memristor is an electronic device that can change its resistance value in response to a voltage and remember previous voltage and current states. A memristor array is a circuit system consisting of many memristors that mimic the synaptic connections between neurons and record the postsynaptic potentials between neurons. The human brain's neural signals then need to be captured. Brain-computer interface technology usually uses electroencephalography (EEG), magnetic resonance imaging (MRI), and other methods to acquire neural signals. These signals are generally weak and require processing, such as signal amplification and filtering, to enhance the strength and accuracy of the signal. After the signal acquisition, data pre-processing is required, including noise removal, filtering, and feature extraction.
These steps can improve the quality and accuracy of the signal and reduce misclassification and interference. The pre-processed neural signals are fed into the memristor array for simulation. In the memristor array, each memristor represents a neuron, and the connection and synaptic strength between them can be regulated employing voltage and current, etc. The memristor array can simulate and record the postsynaptic potentials and signaling between neurons. Finally, the simulation results of neural signals are interpreted and controlled by algorithms and other means. The brain-computer interface system can control external devices such as computers and prostheses by interpreting neural signals such as brain waves and can also realize applications such as human-computer interaction. In the practical application of this system, many details still need to be considered, such as different functional requirements, differences in the source of signal letter acquisition and the environment used, and other detailed adjustments to the application of different scenarios.
WiMi's system adopts the latest technological solution to achieve efficient neural signal processing and analysis by simulating the synaptic connections between neurons through an array of memristors. This technology will lead a new revolution in brain-computer interface technology and bring a more convenient and efficient intelligent interaction experience for human beings. The system also adopts adaptive adjustment algorithms and reinforcement learning algorithms, which can quickly adjust the parameters of the neural network according to the user's operating habits and intentions, thus achieving more accurate control. In addition, the system introduces multimodal sensors and multi-source data fusion technology, which can fuse data from different sensors to improve the accuracy and robustness of the signal.
Artificial intelligence, machine learning, and brain-computer interface technologies are developing rapidly. As an emerging hardware gas pedal with low power consumption, high speed, and high accuracy, memristor arrays have a broad application prospect in neural signal analysis. The neural signal analysis system based on memristor arrays has been verified in the laboratory for several experiments and achieved excellent results. Compared with traditional brain-computer interface technology, WiMi's memristor-based neural signal analysis system has higher efficiency and accuracy and can realize more complex control tasks and interaction modes. The launch of this system will significantly impact healthcare, education, and entertainment, bringing a smarter, more convenient, and more comfortable future for human beings.
StockpickerCAPC
1 year ago
WIMI (NASDAQ: WIMI) Releases Brain-Computer Interface (BCI) Encrypted Anonymizer Software Technology System.
Source
https://finance.sina.com.cn/tech/roll/2023-06-06/doc-imywiftk2574032.shtml
June 6, 2023
Brain-computer interface technology is a cutting-edge technology with great prospects and a broad market application base.
It has unique advantages:
it can access the neural data decoded by the brain in real time. Even sports and other activities are combined to carry out neuromodulation technology. With the rapid development of technology, mobile technology, communication and cloud computing, the brain-computer interface technology has developed rapidly, enabling us to explore things that were unimaginable a few years ago. Application scenarios, the development far exceeds expectations. Currently there are business-oriented (B2B) devices on the market, and consumer-oriented (B2C) commercial devices. Although the development of brain-computer interface technology has brought many opportunities and application prospects, it also brings Here comes some security and privacy concerns.
According to reports, in order to ensure the security and privacy protection of brain-computer interface technology, WIMI (NASDAQ: WIMI) is developing a software system brain-computer interface encrypted anonymizer to ensure user information security and privacy issues.
Brain-computer interface (BCI) technology represents a direct communication link between the brain and external devices. Recent findings demonstrate how electroencephalography (EEG) signals recorded from consumer-grade BCI devices can be used to extract private information about users. With enough computing power, this information can be used by others to infer our memories, intentions, conscious and unconscious interests, and our emotional responses.
This technology has a wide range of applications, from medical devices to gaming and entertainment. However, the use of BCIs raises security and privacy concerns because EEG signals can reveal sensitive information about a user's thoughts, emotions, and memories. Therefore, it is critical to develop techniques that ensure the privacy and security of BCI users.
Research on improving the privacy and security properties of BCI-enabled technologies involves two major steps.
The first step focuses on identifying which components of electroencephalography (EEG) signals can be used to extract private information. After identifying potential vulnerabilities, WIMI's R&D team will quantify the amount of information exposed. Based on the results obtained,
the second step was the development of software encryption tools designed to prevent the possible extraction of users' private information. The "BCI encrypted anonymizer" developed by WIMI is based on signal components, that is, the recorded brain signals can be decomposed into a collection of characteristic signal components in real time. From these components, information corresponding to the user's expected BCI commands can be extracted while filtering out potentially private information.
EEG signals are a combination of different frequencies, each with unique characteristics. The EEG signal components used for private information extraction, the EEG signal components most susceptible to private information extraction are alpha waves and beta waves. Alpha waves are associated with relaxation, while beta waves are associated with cognitive processing and attention. Thus, alpha and beta waves can reveal sensitive information about a user's emotional state, cognitive ability, and attention span.
In order to quantify the amount of information that can be extracted from EEG signals, WIMI WIMI will use the mutual information (MI) method. MI measures the amount of information shared by two variables. In this case, WIMI's R&D team will measure the MI between the EEG signal and the private information that can be extracted from it. Based on MI results, the risk level of extracting private information from EEG signals can be determined.
The BCI encrypted anonymizer developed by WIMI Hologram is a software tool that can prevent the extraction of private information from EEG signals. The tool works by decomposing the EEG signal into a set of characteristic signal components. These components are the building blocks of the EEG signal and can be used to reconstruct the original signal. However, BCI anonymizers filter out components that contain private information, allowing only those needed to control external devices. This technology aims to strengthen the development of closed-loop brain-controlled interfaces, so as to better protect the privacy and security of users. We believe that this technology will provide a safer and more reliable option for future BCI applications.
The brain-computer interface encrypted anonymizer released by WIMI (NASDAQ: WIMI) works in real time without any additional hardware or sensors. It can be integrated into existing BCI systems and customized for different applications. The tool can also be used for offline data analysis to identify potential privacy risks in existing BCI data.
The privacy and security of brain-computer interface technology is critical to ensuring the trust and acceptance of the technology by users in its future development. WIMI's technical solutions can help address security and privacy concerns by identifying EEG signal components that are susceptible to private information extraction and developing software tools to prevent their extraction. The encrypted anonymizer of brain-computer interface can improve the privacy and security of users without compromising the function of brain-computer interface system. This technology could lead to the development of more advanced brain-computer interface systems that provide real-time control of external devices without compromising user privacy. This technical solution can effectively improve the security and reliability of brain-computer interface devices, avoid data leakage and abuse, and provide better user experience and services.
StockpickerCAPC
1 year ago
WiMi Hologram Cloud Launches Its Transformer Algorithm-based WiMi AI Assistant.
Source
https://finance.yahoo.com/news/wimi-hologram-cloud-launches-transformer-120000316.html
June 5, 2023
WiMi Hologram Cloud Inc. (NASDAQ: WIMI), a leading global Hologram Augmented Reality ("AR") Technology provider, today launched an artificial intelligence assistant using an advanced transformer algorithm, called WiMi AI Assistant, which is built up through years of R&D and improvement. WiMi successfully applied the transformer algorithm to natural language processing, providing users with a more intelligent and efficient communication platform.
The WiMi AI Assistant has the following highlights:
- Transformer algorithm:
WiMi AI Assistant uses an advanced transformer algorithm to automatically learn and understand human language and engage in intelligent conversations with humans. The introduction of the transformer algorithm enables the assistant to handle long-distance dependencies and complex contexts better, thus improving overall language understanding.
- Natural Language Processing:
Through deep learning and big data, WiMi AI Assistant understands the user's context and needs and can quickly answer complex questions accurately. In addition, the assistant can learn and improve by itself, continuously optimizing the accuracy and efficiency of question responses.
- Personalized recommendations:
WiMi AI Assistant can provide users with customized information recommendations based on their behavior and interests. Whether it's news, music, or movies, the assistant can precisely meet users' needs and bring them a more personalized experience.
- Multiple applications:
WiMi AI Assistant can be applied in smart homes, intelligent customer service, and many other industries such as healthcare, finance, and education. This has significantly expanded the influence of artificial intelligence.
WiMi Hologram Cloud has always been committed to the R&D and application of innovative technologies. The launch of the WiMi AI Assistant is another significant breakthrough in artificial intelligence. WiMi will continue to increase its investment in R&D to provide users with more high-quality intelligent products and services.
StockpickerCAPC
1 year ago
WIMI (NASDAQ: WIMI) develops a bionic pattern recognition (BPR)-based convolutional neural network (CNN) image classification technology solution.
Source
https://cj.sina.com.cn/articles/view/7651844612/1c815e20402001gjlu?from=finance
June 5, 2023
In recent years, with the continuous development and application of artificial intelligence technology, image classification technology has been widely used in many fields. And with the rise of deep learning, convolutional neural network (CNN) has become the mainstream model for processing image classification tasks. CNNs recognize images by automatically extracting features from them, and classify them using the softmax function. However, due to the limitations of the softmax function, traditional CNN models have some deficiencies in image classification.
It is reported that in order to solve this problem, WIMI has developed a new image classification method that uses a hierarchical structure inspired by the animal visual system to automatically extract features from images. This method combines bionic pattern recognition (BPR) with CNN, which can make full use of the geometric structure of high-dimensional feature space, so as to achieve better classification performance, so it can overcome some shortcomings of traditional pattern recognition. The method has been validated in multiple experiments and, in most cases, achieves higher classification performance than traditional methods.
A Convolutional Neural Network (CNN) is a deep learning model specifically designed to process images. It can automatically extract features from images through convolution and pooling operations, and use fully connected layers for classification. The convolution operation refers to applying a convolution kernel (also known as a filter) to each location on the image and outputting the result as a feature map. The pooling operation refers to downsampling on the feature map to reduce the amount of computation and the risk of overfitting.
In the traditional CNN image recognition classification model, the softmax function is used for classification. The softmax function can convert a set of scores into a probability distribution, where each score represents a confidence score that the image belongs to a certain class. Traditional pattern recognition methods usually use hyperplanes in feature space to segment categories. However, this method has some disadvantages, such as requiring manual feature selection and difficulty in handling nonlinear data. In contrast, bioinspired pattern recognition (BPR) can overcome these problems by performing class recognition on geometric cover sets that are unioned in a high-dimensional feature space.
BPR is a pattern recognition method based on bionics.
Its basic idea is to use biological systems to simulate the processing of sensory information, and regard the pattern recognition process as being carried out in a high-dimensional feature space. In this high-dimensional space, each sample point is regarded as an object rather than a point. Therefore, samples of different classes are distributed in different regions in the high-dimensional feature space, and these regions are called geometric covering sets. Each geometry cover set consists of a set of geometric objects called geometric primitives, such as spheres, cones, polyhedra, etc. With a proper combination of geometric primitives, a coverage set with high classification performance can be constructed to achieve class recognition.
Research shows that WIMI combines BPR and CNN to achieve better image classification effect. Specifically, based on bionic pattern recognition (BPR) convolutional neural network (CNN) image classification, CNN features can be mapped into a high-dimensional feature space, and a geometric coverage set can be constructed in this space, and then the new sample map Go into that space and decide what category it belongs to.
According to the data, WIMI uses a mapping function to map CNN features into a high-dimensional feature space in BPR-based CNN image classification. This mapping function can be a simple nonlinear transformation such as a polynomial transformation or a radial basis function (RBF) transformation. It is also possible to use some more complex function, such as a neural network or a support vector machine (SVM), to learn this mapping function, converting the CNN features into a form that is easier to classify in a high-dimensional feature space.
WIMI Holographic CNN-BPR image classification technology uses proven geometric primitives with high classification performance in high-dimensional feature spaces, such as spheres, cones or polyhedrons, to construct geometric coverage sets. Then, we can use some optimization algorithm, such as genetic algorithm or particle swarm optimization algorithm, to search for the optimal combination of geometric primitives to construct the optimal geometric cover set. Finally, we can use a classifier, such as the K-Nearest Neighbors algorithm or a Support Vector Machine (SVM), to identify the class to which the new sample belongs.
The specific way to realize the image classification method combining BPR and CNN is as follows:
Prepare training and test datasets:
You need to collect a dataset that contains images of many different categories.
This dataset should consist of two parts:
a training dataset and a testing dataset. The training data set is used to train the CNN model, and the test data set is used to test the performance of the classifier.
Train the CNN model to extract image features:
Use the training dataset to train the CNN model and use the model to extract the features of each image. These features will be used to construct a geometric cover set in a high-dimensional feature space.
Mapping CNN features into a high-dimensional feature space:
A mapping function needs to be used to map CNN features into a high-dimensional feature space. This mapping function can be learned using some nonlinear transformation, such as polynomial transformation or RBF transformation, or using more complex functions, such as neural network or SVM.
Build a geometric cover set:
Use some geometric primitives that have been proven to have high classification performance in high-dimensional feature spaces, such as spheres, cones, or polyhedra, to build geometric cover sets. Then, we can use some optimization algorithm, such as genetic algorithm or particle swarm optimization algorithm, to search for the optimal combination of geometric primitives to construct the optimal geometric cover set.
Classify new samples:
Finally, use a classifier, such as K-Nearest Neighbors or SVM, to identify the class to which the new samples belong. We can map the features of a new sample into a high-dimensional feature space, then find the nearest geometric cover set in this space, and finally classify the new sample into the category represented by the cover set.
In addition, WIMI's CNN-BPR image classification technology features a combination of convolutional neural network and bionic pattern recognition, and performs image classification by constructing a geometric coverage set in a high-dimensional feature space. Compared with the current traditional CNN model that uses the softmax function for classification, the softmax function has limited capacity and cannot handle complex classification problems well, such as image classification. In addition, CNN models cannot fully exploit the geometric structure of high-dimensional feature spaces, thus failing to achieve optimal classification performance. And traditional pattern recognition methods usually need to manually select features and classifiers, requiring a lot of manpower and time costs.
By combining BPR and CNN, this technology can overcome some shortcomings of traditional pattern recognition, improve the performance of image classification, and can handle complex image classification problems. This method can overcome some shortcomings of the current traditional pattern recognition in image classification and in most cases, it has higher classification performance than traditional methods. And it can handle complex image classification problems, such as image recognition, object detection and image segmentation.
At present, image classification technology based on convolutional neural network has been widely used in many fields. The method of WIMI hologram combined with bionic pattern recognition can overcome the limitations of traditional pattern recognition methods and improve the accuracy and reliability of image classification. It is believed that with the continuous development and progress of technology, this technology will have wider application and better performance in the future.
StockpickerCAPC
1 year ago
Brain-computer interface major achievements released, WIMI.US has been deeply involved in the development of BCI game models and paradigms.
Source
https://finance.sina.com.cn/tech/roll/2023-06-02/doc-imyvwfif8543218.shtml
June 2, 2023
Last week, Tesla CEO Elon Musk tweeted that the brain-computer interface could solve the biggest bottleneck in human progress. As we all know, Musk launched Neuralink in 2016. He believes that the company's technology can help humans and artificial intelligence achieve a "symbiotic" state. Simply put, people will be able to integrate their brains with computers effortlessly.
When it comes to brain-computer interfaces, the first impression in everyone's mind may be the scene where Musk lets monkeys realize "idea typing". Since then, Musk's Neuralink has been developing brain implants aimed at treating paralysis and blindness, etc. disease. However, what surprised the outside world recently is the team from China.
The world's first "Monkey Brother" interventional brain-computer interface test was successful
Recently, the world's first non-human primate interventional brain-computer interface test was successful in Beijing, which is of great significance to promoting research in the field of brain science and marks that my country's brain-computer interface technology has become an international leader.
The experiment was led by the team of Professor Duan Feng of Nankai University, jointly completed with the General Hospital of the Chinese People's Liberation Army (301 Hospital) and Shanghai Xinwei Medical Technology Co., Ltd., which made breakthroughs in intravascular EEG signal acquisition and interventional EEG signal recognition. Core Technology. Judging from the video data, the post-operative monkey only needs to "think about it" before it can let the robotic arm deliver food into its mouth.
This technology has a wide range of applications in medical, military and other fields. For example, it can help patients with stroke and frostbite recover, and it can even store human thinking, consciousness, and memory in the future. This series of results that sound amazing has already begun to be realized. It seems that the scenes in some sci-fi blockbusters are not so far away from us.
Brain-computer interface refers to the direct connection created between the human or animal brain and external equipment to realize the information exchange between the brain and the equipment.
At present, there are three main types of brain-computer interface, including:
- invasive brain-computer interface,
- non-invasive brain-computer interface and
- interventional brain-computer interface.
Neuralink, founded by "Silicon Valley Iron Man" Musk and a team of scientists, is a representative of a small number of companies engaged in the development of invasive brain-computer interfaces.
In contrast, the interventional brain-computer interface technology used in the Chinese trial is safer. According to the information released by Nankai University, the research team let the interventional EEG sensor pass through the monkey's jugular vein, enter the sagittal sinus, and reach the monkey's motor cortex brain area. After the operation, the team successfully collected and recognized the interventional EEG signals of non-human primates, realizing the active control of the mechanical arm by the animals.
It can be said that the interventional brain-computer interface takes the advantages of both invasive and non-invasive at the same time, while avoiding the disadvantages of both. Judging from the current development, brain-computer interface technology has continuously made breakthroughs, which may first benefit the medical industry. Guotai Junan Securities pointed out that medical health is the main application scenario of brain-computer, and it will gradually penetrate into entertainment, smart home and other fields in the future, becoming an important form of human-computer interaction.
Brain-computer interface technology will empower the metaverse
Brain-computer interface is a complex system and a basic tool for reading and writing neuron dynamic network data. What is even more surprising is that with the sudden emergence of the Metaverse, people have great expectations for the integration and interaction of the virtual world and the real world. Many people in the industry believe that the brain-computer interface is a cutting-edge technology hatched by the Metaverse.
In the future, the brain-computer interface technology will empower the metaverse, and the brain-computer interface is expected to become the next generation of human-computer interaction technology and the ultimate form of the metaverse. At the same time, with the continuous development of the Metaverse, the strong support of national policies, and the continuous exploration of brain science, the brain-computer interface will become the entrance of the next generation of Metaverse after VR and AR, realizing the real Metaverse.
According to the "White Paper on Brain-Computer Interface Standardization" released by the China Electronics Standardization Institute, the potential market for Brain Computer Interface (BCI) technology will soon reach tens of billions of dollars. Another predictive analysis believes that in the next 20-30 years, the commercial application of brain-computer interface will gradually land, which will open up a market worth hundreds of billions of dollars.
The technological race has already begun. Brain-computer interface technology is an important strategic direction for a new round of scientific and technological revolution and industrial transformation. The future industry represented by brain-computer interface has already entered the fast lane, and the formation of breakthrough scientific research results and innovative applications is a strategic need for innovative development.
WIMI Hologram focuses on the field of brain-computer interface and has made great achievements
At present, the field of brain-computer interface has tried to integrate technology in the consumer field, and its application in game interaction is becoming more and more popular. It is understood that WIMI (NASDAQ: WIMI) has started to develop the BCI game model and paradigm based on the brain-computer interface. A game model is designed with a P300 brain-computer interface to explore a feasible and natural game execution experience using electroencephalography (EEG) signals in a real-world environment.
According to reports, the novelty of WIMI's research is reflected in the design of the BCI game and paradigm, which integrates the game rules and the characteristics of the BCI system. In addition, a convolutional neural network (CNN) algorithm is introduced to achieve high accuracy on training samples. This brain-computer interface system not only provides users with a form of entertainment, but also provides more manipulation possibilities for games.
To be sure, WIMI is based on the CNN BCI game model, forming a platform that can satisfy the interests of both healthy and disabled users. For healthy users, brain-computer interface games are mysterious and technical, which increases the charm of the game and is very conducive to the promotion of the game. For disabled users, BCI games provide them with a fair gaming platform, not only allowing them to play games in the same way as healthy users, but also as a functional rehabilitation system to help patients with rehabilitation training. Obviously, WIMI's application of brain-computer interface technology to entertainment games is an important part of promoting BCI technology from the scientific research stage to the actual application market stage.
end
On May 29, Zhao Zhiguo, chief engineer of the Ministry of Industry and Information Technology, introduced that with the joint efforts of the industry, my country has formed a whole industrial chain of brain-computer interface covering the basic layer, technical layer and application layer, and it has been applied in fields such as medical care, education, industry, and entertainment. landing. The Ministry of Industry and Information Technology will take the brain-computer interface as an important direction for cultivating future industrial development, strengthen the exploration of brain-computer interface application scenarios, and accelerate the vigorous development of the brain-computer interface industry.
StockpickerCAPC
1 year ago
WIMI (NASDAQ: WIMI) develops virtual digital human rendering technology based on real-time dynamic simulation.
Source
https://cj.sina.com.cn/articles/view/7651844612/1c815e20402001ghh4?from=finance
May 26, 2023
Virtual digital human refers to a virtual person or avatar created by using computer technology, simulation technology and artificial intelligence technology. Humans interact in various ways such as voice, text and touch. With the rapid development of artificial intelligence, cloud computing, VR/AR and other technologies, virtual digital human technology is also constantly improving, and the virtual digital human industry is currently in a stage of rapid development. More and more application scenarios are emerging, including games, education, medical care, entertainment and other fields. Virtual digital humans have broad application prospects and commercial value. The key technologies of virtual digital human mainly include natural language processing, graphics rendering, multi-modal interaction technology, etc.
The significance of virtual digital human rendering technology to the development of virtual digital human is to improve the fidelity and realism of virtual digital human, so as to make it closer to the expression form of real people, and enhance the user's immersive experience and interactivity. On the one hand, virtual digital human rendering technology can perform fine control in image generation, lighting, shadows, materials, etc., making the appearance and movements of virtual digital human more natural and realistic. On the other hand, virtual digital human rendering technology can be dynamically adjusted according to different scenarios and interaction requirements, thereby increasing the plasticity and personalization of virtual digital human.
It is understood that the research and development team of WIMI (NASDAQ: WIMI) is developing a virtual digital human rendering technology based on real-time dynamic simulation. The virtual digital human rendering technology based on real-time dynamic simulation is a technology for real-time rendering and dynamic simulation of digital characters through computer programs. Using theories and technologies in multiple fields such as 3D computer graphics, dynamics, and biomechanics, real-time calculations and simulated renderings of the appearance, shape, action, and expression of digital human beings are performed to achieve a sense of reality and lifelikeness. A more realistic interactive experience.
The realization of the virtual digital human rendering technology based on real-time dynamic simulation developed by WIMI requires high-precision modeling and dynamic control of digital characters, including the establishment of the skeletal system, the simulation of physical characteristics such as muscles, skin, and hair, and the human body. Calculation of kinematics and aspects of motion capture. At the same time, it is also necessary to use real-time rendering engine, lighting simulation technology, etc. to realize the realistic rendering and dynamic simulation of digital characters in the virtual scene.
The core of virtual digital human rendering technology based on real-time dynamic simulation is real-time dynamic simulation and rendering. The digital human is modeled by computer, and the physical engine is used to simulate the digital human movement, physiological characteristics and the influence of the external environment, and it is rendered in real time to achieve a highly realistic effect.
Specifically, digital human modeling mainly includes:
- human bone structure,
- muscle system,
- skin tissue,
- facial expressions, etc., as well as
- motion capture and motion generation of digital human.
Real-time dynamic simulation uses a physics engine to simulate the physical movement of a digital human, and on this basis calculates the physiological characteristics of the digital human, such as heart rate, breathing, and muscle fatigue, and then simulates the response of the digital human to the external environment. In terms of real-time rendering, the virtual digital human rendering technology uses advanced graphics technologies, such as physically-based ray tracing, material reflection, and panoramic rendering. In this way, a realistic digital human rendering effect can be achieved in a short time, which greatly reduces the pressure on the computer.
According to the data, the virtual digital human rendering technology based on real-time dynamic simulation developed by WIMI has technical advantages such as real-time performance, high fidelity, dynamic interaction and personalized customization. The virtual digital human rendering technology based on real-time dynamic simulation can simulate and render in a real-time environment, so that the movements and expressions of digital human can be displayed in real time.
By simulating details such as human muscles, bones, skin, and hair, it presents subtle and complex movements and expressions more realistically, making digital humans closer to the shape and behavior of real humans, and enhancing the user's sense of immersion and visual experience. It can also be adaptively adjusted according to different scenes and environments, making the performance of the virtual digital human more natural. It can also be customized according to user needs. For example, the appearance, clothing, image, movements and expressions of the virtual digital human can be adjusted to meet personal needs and aesthetics. At the same time, it can also realize the interaction with the user, and respond according to the user's actions and voice.
At present, the virtual digital human rendering technology based on real-time dynamic simulation developed by WIMI Hologram is widely used in games, virtual reality, augmented reality, human-computer interaction, education, advertising and other fields. For example, in film production, the use of virtual digital human rendering technology can significantly improve production efficiency, reduce production costs, and can also create more realistic visual effects for films; in games, this technology is also widely used, making the game experience more Realistic and smooth. In addition, in the field of virtual reality and augmented reality, virtual digital human rendering technology is also widely used, which can achieve a more realistic virtual reality experience and provide users with a more complete sense of immersion. Through the rendering technology of virtual digital human, it can also realize more intelligent human-computer interaction and provide users with a more natural interaction method.
The virtual digital human rendering technology based on real-time dynamic simulation has broad application prospects, and it has become a popular research direction in the fields of computer graphics and artificial intelligence. In the future, with the continuous improvement of related technologies, it will be more widely used and developed.
StockpickerCAPC
1 year ago
WiMi Hologram Cloud Designed A VR-based Remote Collaboration System.
Source
https://finance.yahoo.com/news/wimi-hologram-cloud-designed-vr-120000462.html
May 24, 2023
WiMi Hologram Cloud Inc. (NASDAQ: WIMI), a leading global Hologram Augmented Reality ("AR") Technology provider, today announced the development of a VR-based Remote Collaboration System using communication technologies, and human-computer interaction. The system creates a shared immersive space that can be interacted with, enabling a multi-person remote collaboration work environment that breaks geographical and time constraints and improves collaboration efficiency and productivity.
First, the system uses VR technology to create a shared virtual space and then uses a graphics rendering engine to render the VR scenes to ensure user experience in the virtual environment. The system can transmit the VR scene data of multiple users to the server side for processing and synchronize the actions and operations of different users to the server side through data synchronization and present them on the terminals of other users. Users can transmit voice, gesture, and other data to other users through the network. After recognizing the user's actions and commands, the system can update the state and scene of the virtual world in real-time to achieve collaborative and natural interaction.
In addition, the VR interaction interface enables the assignment and execution of joint tasks and the sharing of resources. Multiple users interact with 3D objects in a collaborative AR environment to understand each other's completed work and ultimately complete collaborative tasks. Immersive collaboration platforms are more expressive than 2D video remote platforms and have the advantage of being able to share common goals accurately. When using a full-body avatar to collaborate realistically in a virtual environment, the sense of physical ownership is enhanced, and immersion is increased.
VR remote collaboration systems can eliminate time and space constraints and improve collaboration efficiency. Users can communicate in real-time in virtual space, share information and resources, and work more efficiently. In addition, virtual reality technology uses stereoscopic vision and seamless interaction to provide a more realistic user experience. This is something that traditional remote collaboration tools cannot include. VR remote collaboration system is an innovative model of modern remote collaboration that can be more widely used in the future.
The application and market prospect of VR remote collaboration systems is extensive. In the industrial field, this system can help companies to conduct remote collaboration and training, improving productivity and employee skill levels. In the medical area, it can support medical experts in conducting remote consultations and surgical operations and solve the problem of geographical limitation. It can also give students a more prosperous online learning environment and interactive experience to improve teaching quality. And in the market, as people become more and more interested in virtual reality, the application of this technology will become more and more widespread.
StockpickerCAPC
1 year ago
WiMi Hologram Cloud Develops A Human-Robot Interaction System Based on Machine Learning Algorithms.
Source
https://finance.yahoo.com/news/wimi-hologram-cloud-develops-human-120000844.html
May 23, 2023
WiMi Hologram Cloud Inc. (NASDAQ: WIMI), a leading global Hologram Augmented Reality ("AR") Technology provider, today announced the development of an HRI system based on machine learning algorithms. The multi-modal HRI system fuses voice and gestures, which converts the user's voice and gestures into commands that the robot can execute. Using gestures for HRI is an ideal way to do so. This is because gestures express rich semantics and are easy to recognize. Voice interaction based on natural language understanding is the most direct and convenient.
The HRI system, based on machine learning algorithms, uses a combination of gestures and voice to control the robot. Voice is used as a natural interaction method to control the robot, and gestures are used as a complement to voice to improve the accuracy of commands. Combining gestures and speech reduces the disadvantages of using gestures or speech alone and makes communication between humans and robots more natural, efficient, and accurate.
Through voice interaction, the robot understands what people say and communicates with humans with emotion. The human-robot dialogue system presents humanized and intelligent interaction characteristics. Gesture recognition is based on the motion trajectory of human hands and simulates images or syllables according to the change of gestures. Thus, certain meanings or words are formed to express thoughts vividly, allowing the robot to understand and interact with human language.
With the gradual maturity of HRI and the application of voice emotion recognition in people's lives, the need for machine intelligence to understand human emotions has become more urgent.
WiMi's HRI system provides a faster, more efficient, and more diverse interaction experience by fusing multi-modal perceptual information. Using gestures and voice for real-time parallel interaction, visual communication, and voice information are associated and shared in real-time during the interaction. Multiple interaction modes complement each other to form a complete interaction system. The system leads the gradual development of HRI in the direction of intelligence and humanization and helps build a harmonious and natural human-robot environment.