StradVision announces its software will be featured in multiple mass-production ADAS and self-driving vehicle projects, with cars in production in 2019
COO Sunny Lee touts SVNet’s capabilities, company’s aggressive timeline at Automotive Innovation Summit
StradVision COO Sunny Lee announced at the KPMG/Flex Automotive Innovation Summit in Milpitas, CA, that the company’s SVNet software will be used in five production vehicle projects with OEM partners in China and Germany, with the first vehicles in production in 2019.
StradVision is a vision processing software company, offering AI perception solutions critical for the development of autonomous vehicles.
Its core technology is a deep-learning camera software called SVNet. StradVision currently provides Level 2-certified technology for ADAS systems. By 2021, StradVision plans to have more than 6 million vehicles on the road using SVNet, which is already compliant with strict standards such as Euro NCAP and Guobiao (GB) in China.
Beyond 2021, StradVision will provide software for Level 4 autonomous vehicles.
StradVision also earned the ASPICE CL2 certificate, which recognizes suppliers with a serious, well-managed software development process who ensure safety and accuracy during product development and implementation.
“Some of our competitors may have flashier products, but they cannot match the level of real-world application that StradVision will realize in years to come,” Lee said. “We are goal-oriented, offer quick execution and excel at implementing Deep Neural Network-based object detection on embedded platforms within a short period, approximately three to six months.”
StradVision’s SVNet External enables vehicles to execute ADAS and self-driving functions, including:
- Forward Collision Warning
- Traffic Sign Recognition
- Intelligent Speed Assistance
- Pedestrian and Cyclist Warning
- Pedestrian Collision Warning
- Lane Departure Warning
- Blind Spot Detection
SVNet Internal monitors driver and passenger to ensure a safe driving experience. Functions include Gaze Detection, Drowsiness Detection and Age & Gender Detection. It works with platforms including Renesas R-Car, NXP S32V, Qualcomm Snapdragon, Nvidia TX2 and DPX2, and Texas Instruments. The algorithm allows both internal and external technology to work at the same time.
StradVision’s SVNet Tools enables operational efficiency by guaranteeing data independence. Functions include Auto Labeling System, Data Training Suite and Platform Optimization Suite.