In a field as multidisciplinary as AI, collaboration is often necessary for the best results. For example, 3D vision relies on the culmination of expertise in AI hardware, vision sensing, software architecture, and model development and training. For this reason, companies tend to work together to develop AI solutions, with Nvidia being the industry’s leader in AI hardware.
Recently, Orbbec announced the release of the Persee N1: a new camera-computer kit designed for 3D vision applications. The new product leverages Nvidia’s Jetson platform as its computational backbone.
Front view of the Persee N1. Image used courtesy of Orbbec
The Persee N1
The Persee N1 is an all-in-one camera-computer kit designed for 3D vision applications. The module consists of two major components: a computing platform and a depth camera.
The core of the Persee N1 is the Nvidia Jetson Nano platform. This platform consists of a Quad-Core Arm A57 processor, clocked at 1.43 GHz, along with a 128-core Nvidia Maxwell architecture GPU. Together, these resources yield an AI performance of 472 GFLOPS.
The camera component of the Persee N1 is Orbbec’s Gemini 2 camera, based on active stereo IR technology. This technology, combined with Orbbec’s custom ASIC, ensures high-quality depth processing. The depth camera offers a wide field of view (H91 degrees x V66 degrees) and a depth range of 0.15–10 meters. It can also deliver a depth resolution of up to 30 frames per second at 1280 x 800 and an RGB resolution of 1920 x 1080.
Furthermore, the device supports multi-camera synchronization and includes an inertial measurement unit (IMU), enhancing its capabilities for complex applications. The Persee N1 operates effectively in indoor and semi-outdoor environments, with an operational temperature range of 0°C to 40°C.
Orbbec developed the Persee N1 system for medtech, dimensioning, interactive gaming, and robotics applications.
Plugging In the AI Computing Power of Nvidia Jetson
Nvidia designed its Jetson platform to accelerate AI performance for edge computing devices in a power-efficient and compact form factor. Using Nvidia’s CUDA-X software, the platform supports cloud-native technologies like containerization and orchestration, enabling the development and deployment of AI applications at the edge.
Nvidia Jetson modules. Image used courtesy of Nvidia
The family includes several options, each catering to a range of performance levels and form factors in edge computing. Some key products in the Jetson family include the Jetson AGX Orin Series, the Jetson Orin Nano Series, the Jetson Orin NX Series, the Jetson Orin Nano Series, the Jetson Xavier NX Series, and the Jetson Nano.
Orbbec claims that the combination of the Nvidia Jetson platform with the company’s Gemini 2 RGB-D camera opens doors for more advanced edge AI and robotics while also “enabling large-scale cloud-based commercial deployments.”
A Collaboration in AI Development
Nvidia wasn’t the only external party to support Orbbec’s new camera-computer kit. Open Computer Vision (OpenCV), run by the non-profit Open Source Vision Foundation, also pledged support for the Persee N1 with integration to its popular OpenCV library. This library of programming functions for real-time computer vision is used in approximately 89% of all embedded vision projects, according to the company.
By leveraging Nvidia’s Jetson platform and the OpenCV library, Orbbec integrated specialized AI hardware and software into its 3D vision solution and made a more comprehensive vision solution available to the end user.
Trending Products
