StradVision Showcases the Most Advanced Camera Perception Technologies at AUTO TECH 2021

StradVision’s deep learning-based perception software SVNet enables the most innovative ADAS features such as Highway Driving Assist and Automated Valet Parking

StradVision, a pioneer in AI-based vision processing technology for Autonomous Vehicles and ADAS systems, has announced that it will showcase the latest features of its camera perception software SVNet during The China Guangzhou International Automotive Technology Expo 2021 (Auto Tech), from the 25th to the 27th of May. 

Auto Tech which will be held at Guangzhou Poly World Trade Center Exhibition Hall in China, covers all the important topics of the automotive industry such as automotive electronics, connected cars, EV&HEV, autonomous driving technology, etc. Over 500 global top companies, including OEMs, auto research institutes, and Tier 1 suppliers, will show their latest technologies and products for the automotive industry at the event. 

At booth #C357A, StradVision will demonstrate its Front-Facing Camera solutions based on NVIDIA Xavier SoC. In addition, it will expand its reveal to the latest technologies such as Surround View Monitoring and Pseudo Lidar, which are attracting great attention in the industry. 

The key disclosures include: 

  • Introducing the improved perception capability in longer-distance by higher resolution images of 8 megapixels (3840×2160), and the ground-breaking Highway Driving Assist (HDA) feature implemented based on this.  
  • Unveiling Multi-camera 360-degree perception using up to 9 cameras, which is critical to implement autonomous driving features of L3 or above, such as Automated Valet Parking (AVP), and Enhanced Autopilot. 
  • Innovative Pseudo Lidar technology that enables depth estimation of objects by mono channel camera without high-cost and high-performance lidar equipment. 

SVNet, a deep learning-based perception software 

SVNet is a lightweight software that allows vehicles to detect and identify objects accurately, such as other vehicles, lanes, pedestrians, animals, free space, traffic signs, and lights, even in harsh weather conditions or poor lighting. 

The software relies on deep learning-based perception algorithms, which compared with its competitors is more compact and requires dramatically less memory and electricity to run. SVNet supports more than 14 hardware platforms and can also be customized and optimized for any other hardware system thanks to StradVision’s patented and cutting-edge Deep Neural Network-enabled technology. 

SVNet is currently used in mass production models of ADAS and autonomous driving vehicles that support driving automation levels 2 to 4, and will be deployed in more than 13 million vehicles worldwide. 

Recent contract wins for the next-gen ADAS technologies 

StradVision recently partnered with a leading German automotive OEM to supply augmented reality to the Navigation and Lane Keeping Assistance Systems (LKAS) of its vehicle lineup. Additionally, it began working with a global Tier 1 automotive supplier to provide a Surround View Monitoring (SVM) algorithm that supports Park-Assist functions such as Automatic Parking Assistance (APA).  

StradVision’s augmented reality technology will be featured on more than 40 vehicles from the OEM, scheduled for production globally in 2022. The augmented reality technology will be powered by StradVision’s SVNet software, which powers an enhanced depth-map solution, semantic segmentation, and lane detection to improve visuals for the center screens. 

Also, StradVision’s software contributions to the SVM partnership will ensure that the camera system can be used for surrounding objects and free space detection to support multiple features, including Reverse Brake Assist, and Automated & Remote Park Assist function. 

Join Our Newsletter!

Subscribe to our newsletter to receive the latest news

Please enter your name.
Please enter a valid email address.
Please check the required field.
Something went wrong. Please check your entries and try again.