CNN

The computational and algorithmic demands made by computer vision systems highlight HLS' value for AI system development.

Machine-learning strategies for embedded vision are evolving so quickly that designers need access to flexible, heterogenous processor architectures that can adapt as the algorithms evolve.

Many car manufacturers are exploring the possibilities of autonomous vehicles. But what will it take to build sufficient AI performance into them to enable true autonomy?

High-performance vision-processing algorithms need optimized CNN engines to deliver the right performance within the power budget of embedded applications.

Using specialised processors to implement key AI computation tasks such as CNNs.

Quadrupling the performance of a dedicated CNN engine within an embedded vision processing core brings more complex graph processing within reach.

Using deep learning techniques and convolutional neural networks to bring facial recognition capabilities to embedded systems.

Enabling autonomous driving will demand embedded processors that can process multiple HD video streams and analyse them using convolutional neural networks.