The KUMA200 is the new extreme wide-angle multi-camera AI sensor from the Belgium-based company Metal. The Canaan’s Kendryte K210 dual-core RISC-V processor with a neural network coprocessor powers a compact multi-camera sensor.
The AI sensor’s design intent is for low-power artificial intelligence applications. Some of these applications include face detection, object recognition, and audio processing at the edge. The KUMA200 powered by the Kendryte K210 RISC-V chip is designed to unite four cameras on one ultra-compact device.
What’s so special about the Kendryte K210 Microcontroller?
Canaan’s K210 MCU includes a KPU, a self-developed neural network hardware accelerator capable of high-performance convolutional neural network operations. So in terms of AI computing, K210’s processing capability is pretty outstanding. Additionally, object recognition, video playback, imaging, 3D rendering, and even playing FC simulators on it are just a few of the applications that K210 excels in.
In terms of security, K210 employs an Advanced Encryption Standard (AES) hardware accelerator and a One Time Programmable (OTP) ROMSHA256. Moreover, typical application scenarios consume less than one watt of power and the chip’s power consumption is less than 300 mW.
Kuma200 AI camera’s specifications
The processor module, camera module, 4x OV2640 camera modules, and a 2.4″ QVGA LCD are the components in the Kuma200 sensor kit. It features an 8bit MCU LCD 24P 0.5mm FPC connector with 4x GPIOs connected to the header. In addition, Kuma200 provides a micro-SD card for external storage. RST button and the USR button are also provided on the Kuma200.
Furthermore, this tiny board has USB connectivity and four 24-pin FPC connectors for the camera sensors, with Wi-Fi available “on request” as an optional extra. The UFO-looking camera mount depicted in the company’s graphics is purely for illustration purposes. Xetal, the belgium-based manufacturer is offering buyers CAD files to build their own.
Priced at $68, the Kuma200 AI camera can be availed at Xetal’s Tindie store.
Jennifer James is a graduate student in Computer Science Engineering who is passionate about front-end development. She is a content-writer inquisitive about technology. A rising enthusiast in search for optimum knowledge through learning and experiences of the everyday fast-growing Digital Industry through organizational exposure.
