Jetson Nano and Intelligent Video Analysis

Jetson Nano and Intelligent Video Analysis

Jetson Nano development kit measures only 80x100mm and features four high-speed USB 3.0 ports, MIPI CSI-2 camera connector, HDMI 2.0 and DisplayPort 1.3, Gigabit Ethernet, M.2 Key-E module, MicroSD card slot, and a 40-pin GPIO header. The Nano’s interfaces and GPIO header can connect to various popular peripherals and sensors.

Jetson Nano and Intelligent Video Analysis

The development kit boots from a MicroSD card and can be conveniently powered via a micro USB port or a 5V DC barrel jack adapter.

The camera connector is compatible with affordable MIPI CSI sensors, including the 8MP IMX219 module provided by Jetson ecosystem partners. The Raspberry Pi v2 camera module is also supported, with drivers available in JetPack.

The table shows the key specifications.

Jetson Nano and Intelligent Video Analysis

* indicates the maximum concurrent streams to total throughput.

Supported video codecs: H.265, H.264, VP8, VP9 (VP9 decoding only)

Jetson Nano can process up to 8 HD full-motion video streams in real-time, serving as a low-power intelligent video analysis platform for network video recorders (NVR), smart cameras, and IoT gateways. NVIDIA’s DeepStream SDK utilizes ZeroCopy and TensorRT to optimize the end-to-end inference pipeline, achieving ultimate performance on edge and on-premises servers. The video below shows Jetson Nano performing object detection on 8 streams of 1080p30, running a ResNet-based model at full resolution with a throughput of 5 million pixels per second (MP/s).

The video shows Deepstream running on Jetson Nano, executing a ResNet-based object detector on 8 independent 1080p30 video streams.

Jetson Nano and Intelligent Video Analysis

The diagram above shows an example architecture of an NVR using Jetson Nano, which ingests and processes up to 8 digital streams over Gigabit Ethernet using deep learning analysis. The system can decode H.264/H.265 at 500 MP/s and encode H.264/H.265 video at 250 MP/s.

The DeepStream SDK for Jetson Nano is scheduled for release in Q2 2019.

Scan the QR code to join the NVIDIA DeepStream Developer Program to receive notifications about NVIDIA’s upcoming releases.

Jetson Nano and Intelligent Video Analysis

More about Jetson Nano

Oh no, why isn’t my Jetson Nano powering on?

Finally, the Jetson Nano flashing tutorial video you wanted is here!

Introduction to NVIDIA’s open-source autonomous robot kit JetBot

Several issues to note when purchasing the Jetson Nano development kit

Read the review report of NVIDIA Jetson Nano from foreign media

Jetson Nano and Intelligent Video Analysis

Leave a Comment

Your email address will not be published. Required fields are marked *