Compound Eye, a developer of software for 3D sensing and perception, has made a development kit available for its VIDAS product. The company explains that VIDAS generates dense 3D outputs and per-pixel semantic class across a diverse set of on-road and off-road environments using only scalable, passive sensors like automotive-grade cameras, removing the need for lidar or radar.
The company states that this system enables vehicles to understand their surroundings in dense 3D and in real time, with applications ranging from in-dash displays to driver assistance and autonomy. The software runs on an off-the-shelf embedded computer, delivering what is said to be 10 times the resolution of lidar at a ready-to-scale cost.
Compound Eye states that its system is built atop industry standard hardware and is designed with OEMs and Tier 1s in mind. For ease of packaging and integration, cameras can be mounted almost anywhere on the vehicle. It notes that VIDAS is self-calibrating, can be adapted to work with a wide variety of automotive cameras, and works alongside OEM or third-party software.
The early access VIDAS DevKit includes two reference cameras, a compute module, a GPS antenna, two FAKRA cables, and a 12V power cable. Initial installation takes 30 minutes or less and perception outputs can be viewed immediately on a tablet or laptop. The SDK also includes VIDAS Operator for viewing and recording data and VIDAS Inspector for analyzing perception outputs.