CUDACasts Episode 21: Porting a simple OpenCV sample to the Jetson TK1 GPU

In the previous CUDACasts episode, we saw how to flash your Jetson TK1 to the latest release of Linux4Tegra, and install both the CUDA toolkit and OpenCV SDK.  We’ll continue exploring the power efficiency the Jetson TK1 Kepler-based GPU brings to computer vision by porting a simple OpenCV sample to run on the GPU.  We’ll explore computer vision further in a future CUDACast when we look at the VisionWorks toolkit from NVIDIA.

CUDACasts are short how-to screencast videos about new features and techniques for GPU programming. Click here for all CUDACasts.

To suggest a topic for a future episode of CUDACasts, or if you have any other feedback, please use the contact form or leave a comment below.

  • Chim Lee Cheung

    The last demo is optical flow? Is it running in real time?

  • Augustine Wong


    Can you have a future cudacast in which you demonstrate how to perform cross compilation with a Jetson? Many of the online resources I found describe how to install a cross-compilation platform on the host system but don’t walk through the entire process of installing the cross-compilation platform, compiling CUDA programs on the host, and placing those programs onto Jetson.