Stereo vision systems give the robots depth perception skills, which make artificial machines and systems develop an understanding of their environment by estimating the relative distance of objects in their vision from many visual cues. Hence, stereo vision is used in many areas of robotics, such as self-driving cars, drones for rescue missions, and robots for remote surgery.
Arducam released this Stereo Camera MIPI Module Series for Raspberry Pi and Jetson Nano/Xavier NX. They directly connect to the MIPI CSI-2 connectors of the motherboards and run with a V4L2 camera driver on those platforms. They offer better flexibility to be integrated into your own hardware design or run with your own algorithm on embedded systems for applications like depth sensing, 3d mapping, SLAM, etc.
This Arducam 2MP Stereo Camera MIPI Module is a stereo camera module with two synchronized monochrome global shutter OV2311 image sensors (2x2MP). The monochrome sensor's capability of excellent detail and sensitivity allows it to get higher accuracy and frame rates in extracting depth information.
VR / AR application, etc.
Complete Synchronization(The frames of two cameras are synchronized at the hardware level, extra synchronization scheme by time stamping or so does not need to be considered).
Mono Global Shutter: Shoot high-speed moving objects in crisp sharp images.
The ISP processing function is supported on the Raspberry Pi platform, and the max frame rate is: 50fps@3200×1300
The ISP processing function is not supported on the Nvidia Jetson platform, and the max frame rate is: 50fps@3200×1300
Raspberry Pi and Jetson Nano
x1 CSI-2 2-lane Output Port/1.0 Gbps/Lane
Supported on Raspberry Pi, not supported on Nvidia Jetson
Raspberry Pi & Jetson Nano/Xavier NX boards, acrylic case, and tripod are not included in the package.
This camera relies on kernel drivers to work – one kernel version, one camera driver. Up or downgrading your OS leads to a kernel bump, and you may need to re-install a matching driver.
The driver of the kit is mainly developed based on the Raspberry Pi/Nvidia Jetson official board, and there is no guarantee that the third-party board can be used (due to the different allocation of hardware resources, it is not compatible with the driver)
The I2C signal and clock of each camera are fully synchronized. It is unavailable to capture the same frame of images at different exposure times/gains for each camera.
The camera kit does not have an onboard ISP, using the ISP function is platform-dependent. (On the Raspberry Pi platform, you can use the ISP processing of images. On the Jetson platform, image capture tools and applications conforming to the V4L2 standard framework are supported, but ISP processing of images is not supported).
Our main business is to assist customers in acquiring data from binocular cameras. Due to the extensive backend processing involved in depth vision, we regret that we cannot provide comprehensive operational guidance to cover all requirements. If you have relevant needs, you may refer to open-source demos for binocular cameras, such as those offered by NVIDIA.