You can use Linux for Tegra software to customize your Jetson DevKit platform software to fit the needs of your development project. By following NVIDIA’s development and porting guidelines, you can optimize your use of the complete Tegra feature set. Follow the links below for details about the latest software libraries, frameworks and source packages.
Find the right software for your needs. Our repository has the latest versions of compatible software available for immediate download, and new updates are added on a regular basis. Follow the links below to get started.
The NVIDIA VisionWorks toolkit is a software development package for computer vision (CV) and image processing. VisionWorks™ implements and extends the Khronos OpenVX standard, and it is optimized for CUDA-capable GPUs and SOCs enabling developers to realize CV applications on a scalable and flexible platform.
The NVIDIA CUDA Deep Neural Network library (cuDNN) is a GPU-accelerated library of primitives for deep neural networks. Deep learning developers and researchers worldwide rely on the highly optimized routines in cuDNN so they can focus their efforts on designing and training neural network models rather than spending time on low-level performance tuning.
This is a version of OpenCV that NVIDIA optimized specifically for the Tegra platform family. It is available for Linux for Tegra, Android, and Vibrante (Embedded Linux) operating systems. It contains Tegra-specific optimizations that enable it to often run several times faster than the OpenCV for Android implementation.
Software tools for Linux for Tegra are available under Embedded->Develop->Tools.