WiMi Hologram Cloud Inc (WIMI)

[StockpickerCAPC]

WiMi Hologram Cloud Developed A Intelligent Driving System.

Source
https://www.newstrail.com/wimi-hologram-cloud-developed-a-intelligent-driving-system/

December 13, 2022

Under the trend of integrating artificial intelligence in automobiles, artificial intelligence technology based on neural networks, deep learning, and other algorithms has become a hot and challenging point in the research field of the intelligent driving industry.

With the development of auto driving, pedestrian and obstacle recognition is becoming increasingly important. The research and development team of WiMi Hologram Cloud, Inc. (NASDAQ: WIMI) has developed a pedestrian and obstacle recognition system based on vehicle multi-source information perception technology. This system integrates multi-sensor information such as environment sensing sensor, inertial measurement unit, IMU fusion, and laser radar for intelligent recognition, and carries out real-time dynamic estimation and tracking of distance and speed of objects through millimeter-wave radar and forward-facing camera, thus realizing fast and accurate identification of obstacles and pedestrians.

The WiMi Hologram Cloud multi-source information vision technology uses multiple cameras to collect information from different directions and distances. At the same time, this data is processed, fused, and output to the controller to achieve accurate identification and positioning of pedestrians and obstacles. Based on traditional multi-sensor technology, the depth learning algorithm is further introduced, which can better realize the effective collaboration of physical parameters such as distance, speed, and orientation between vehicles and pedestrians and achieve the comprehensive evaluation and detection of multiple different objects (such as human shape, animal, car, etc.). This method can effectively evaluate the accuracy of vehicle crash analysis and evaluation, reduce the incidence of road traffic accidents, and thus improve the safety of automatic driving.

Automatic driving requires a large number of high-quality, safe, and unbiased data for simulation training, so higher requirements are put forward for the accuracy of data annotation. WiMi Hologram Cloud R&D team is developing artificial intelligence data annotation algorithm technology and applying it to holographic vehicle multi-source information perception pedestrian and obstacle recognition systems. In the process of data preprocessing, synthetic intelligence data annotation is indispensable because the supervised machine learning model can learn to identify the repeated patterns in the annotated data. After processing many labeled data, the algorithm can recognize the same way when new, unlabeled data appears. WIMI Hologram Cloud R&D team has conducted a lot of in-depth research and development in the field of computer vision (boundary box labeling, 3D point cloud labeling, essential point labeling, line labeling, 2D/3D fusion labeling, target tracking, image classification, semantic segmentation, etc.), voice engineering (voice cutting, ASR voice transcription, voice emotion determination, voice print recognition, etc.), and natural language processing (OCR transcription, text information extraction, NLU sentence generalization), independent research and development of data engine framework, data cleaning and annotation through AI test model, and completion of artificial intelligence model training and deployment.

Data annotation is the process of adding metadata to the training dataset. This metadata is usually in the form of tags and can be added to any data, including text, images, and videos. Adding high-quality and high-precision labels is a key process in developing training data sets for machine learning.

WiMi Hologram Cloud, Inc. (NASDAQ: WIMI) introduces artificial intelligence data annotation algorithm technology into the pedestrian and obstacle recognition system of holographic vehicle multi-source information perception technology, which can more effectively improve the efficiency and accuracy of data processing in the field of automatic driving. In the future, AI data annotation will be widely used in more areas, such as medical imaging analysis used to improve disease prediction, diagnosis, and treatment; Face recognition payment, AR/VR, urban security and monitoring, etc.