Today NVIDIA released JetPack 3.1 which introduces L4T 28.1 with production support for both the NVIDIA Jetson TX1 and Jetson TX2 Development Kits. Also new is TensorRT 2.1, cuDNN 6.0 and expanded multimedia API functionality and samples. Components for Jetson TK1 remain unchanged.
JetPack 3.1 is available on the NVIDIA Embedded Developer Website.

From the JetPack 3.1 Release Notes
Release Highlights
- New L4T Production Release 28.1
- This 64-bit BSP (Board Support Package) has been designed to work on both Jetson TX2 and Jetson TX1
- TensorRT 2.1
- New Customer Layer API enables integration of novel, user-defined layers
- Doubled Deep Learning inference performance for batch size of one
- cuDNN v6.0
- New fused convolution provides better performance due to faster compute in the fused kernels
- New dilated convolution reduces the number of parameters, which results in speed up of computation for certain applications like object detection and image segmentation that require convolution followed by upscaling
- Multimedia API v28.1
- New functionality
- TNRv2 (Temporal Noise Reduction)
- High quality spatio-temporal noise reduction using GPU. Recommended for applications where low light video quality is important and GPU requirement is acceptable. Typical GPU utilization is <8.5% for 1080p30fps operation on Jetson TX1.
- Piecewise linear WDR Support
- ISP now supports cameras with “built-in WDR” that combine multiple exposures on-sensor and transmit the result with a piecewise linear encoding. Functionality verified using Sony IMX-185 (and reference driver is included). This feature does not include support for other WDR technologies such as DOL or spatial interleaving.
- New samples
- How to share CUDA buffer with v412 camera and then process color conversion (YUYV to RGB) with CUDA algorithm
- How to render video stream (YUV) or UI (RGB) with Tegra DRM (Direct Rendering Manager), i.e. rendering support for non-X11 and lightweight display system. Tegra DRM is implemented in user-space and is compatible with standard DRM 2.0
- New functionality
JetPack 3.1 is available on the NVIDIA Embedded Developer Website. Go grab some new JetPack goodness!
8 Responses
Hi Kangalow,
Although it is relatively a small part of this new release, I’m interested in the WDR part. Where you able to find additional details about how to use this feature? I checked the Jetpack 3.1 for applications to support it. But couldn’t find any.
The kernel source gives little information other than what is present in the sensor. No documentation or anything as such.
I don’t know anything about that particular feature, but it sounds pretty cool! There is a lot of documentation out there: https://developer.nvidia.com/embedded/downloads
You can certainly ask questions about it in the NVIDIA developer forums: https://devtalk.nvidia.com/default/board/139/jetson-embedded-systems/
Where there’s a wide range of developers, vendors and NVIDIA engineers. Thanks for reading!
I already did.
https://devtalk.nvidia.com/default/topic/1019385/hdr-camera-in-jetson-tx2/
And I did check the provided documentation for any info about it. Couldn’t find any. I was hoping you could shed some more light on this. But thanks anyways. Keep up the good work!
Hi Kangalow,
Have you had success with your ZED camera after installing Jetpack 3.1? I’m receiving an error pointing to line 301 of scan_app.cu. I’m assuming some incompatibility with libcudpp.so prepackaged with the ZED. I sent a note to Stereolabs, but I’m wondering if you’re experiencing the same issue.
By the way, your work is tremendously helpful. Prior to Jetpack 3.1 I was able to modify dusty-nv’s detectnet app to accept the Zed video stream, and once an object was identified obtain the average distance to the object. This was thanks in large part to your various tutorials.
Thanks!
Thank you for the kind word, using the ZED with detectnet sounds like fun! I have not worked with the ZED camera and 3.1 yet. My guess would be that there is a compatibility issue with the new CUDA 8.0 release, it’s not unusual for some of the calls to change. Thanks for reading!
Since the new JetPack has cuDNN 6.0, was anyone successful in installing Tensorflow?
hi dear kangalow
i’m new to jetson tx1
i want to use new driver on tx1
do you know where I could find the BSP soruce!??
https://developer.nvidia.com/embedded/downloads