GTC Silicon Valley-2019: GstCUDA: Easy GStreamer and CUDA Integration
Note: This video may require joining the NVIDIA Developer Program or login
GTC Silicon Valley-2019 ID:S9391:GstCUDA: Easy GStreamer and CUDA Integration
Daniel GarbanzoHidalgo(RidgeRun),Michael Grüner(RidgeRun)
We'll discuss GstCUDA, a framework developed by RidgeRun that provides an easy, flexible and powerful integration between GStreamer audio/video streaming infrastructure and CUDA hardware-accelerated video processing. We'll provide demos and examples that show how to easily develop custom GStreamer elements that execute CUDA algorithms, ensuring zero memory copy pipelines for minimum latency and maximum performance on Jetson platforms. We'll also provide an overview of GstCUDA's key features, which include easy CUDA algorithm integration into GStreamer pipelines, complexity abstraction on both CUDA and GStreamer, out-of-the-box quick prototyping GStreamer elements for different processing topologies, and zero-memory copy integration for increased performance on Jetson boards (up to dual 4K@60fps).