Bookmark and Share

Intel Demonstration of Deep Learning-based, GPU-Accelerated Object Recognition

Jingyi Jin, software architect in the Visual and Parallel Computing Group at Intel, demonstrates the company's latest embedded vision technologies and products at the May 2016 Embedded Vision Summit. Specifically, Jin demonstrates Intel Graphics being used as an accelerator to support real-time object recognition on low-end devices like consumer laptops and tablets. Intel Graphics is pervasive on client platforms, and when well utilized, it can boost performance of a CNN-based application by factor of 5x compared with a default Caffe CPU + MKL implementation on the same device ("Cherry Trail" Atom x7-z8700, for instance). The demonstrated application is built on top of clCaffe (the Caffe framework extended with OpenCL), which is open sourced for easy reproducibility; its OpenCL code path is performance-optimized for Intel Graphics (https://github.com/01org/caffe/tree/clcaffe).