Collaboration on smart camera development for ADAS systems
The new solution is designed to provide high-precision object recognition for detecting vulnerable road users (VRUs) and consumes very low power.
StradVision’s deep learning-based object recognition software has been optimised for Renesas R-Car automotive SoC products R-Car V3H and R-Car V3M. The R-Car devices use a dedicated engine for deep learning processing called CNN-IP (Convolution Neural Network Intellectual Property), allowing them to run StradVision’s SVNet deep learning network at high speed with low power consumption. The resultant solution is suitable for mass-produced vehicles.
StradVision’s SVNet deep learning software can recognise objects precisely in low-light environments and deal with occlusion when objects are partially hidden by other objects. The basic software package for the R-Car V3H can recognise vehicles, person and lane recognition simultaneously at 25 frames per second. The package can be used as a foundation for developers to further customise.
More information
Related news
Complete C/C++ compiler and debugger toolchain for RISC-V
IAR Systems extends Visual State with Java and C# support
Multi-gigabit automotive communication requires a new standard
Renesas R-Car gets Android support package
If you enjoyed this article, you will like the following ones: don't miss them by subscribing to :