SVNet enables external vehicle perception via 14+ platforms, accurately and safely detecting and recognizing objects. Also, it was the first network to run deep learning-based object detection software on TDA2x, V3H, V3M, and H3.
HARDWARE SPECIFIC OPTIMISATION
SVNet has minimum computational requirements and electricity consumption due to its lean design. It is also a hardware optimized solution that can service more than 14+ platforms. Whether the target platform is already on the market or R&D phase, StradVision can make SVNet fully operational with virtually no downtime. SVNet was the first Deep Learning-based algorithm ever to be ported onto TI TDA2x and Renesas' V3H, V3M, and H3.
V2H, V3H, V3M
TDA4x, TDA2x, TDA2Px
Xavier, TX2, DPX2
ZU9, ZU7, ZU5
There are more than three dozen functions that can be executed by SVNet. It also can be customized to detect and perceive various additional objects depending on our clients' needs. This is all possible due to SVNet's flexible and powerful nature.