Français
Deutsch
Español
Italiano
Português
日本語
한국어
Русский
Drop Nineteens, Offset & Cardi B, & The Week's Best New Songs: Listen
Aug 08, 2023Electro Scan Inc. wins Best of Sensors 2023 award for Swordfish
Aug 22, 2023Drone defense firm DroneShield lands $33M U.S. government deal
Aug 18, 202310 Best LED Strip Lights of 2023
Jul 31, 2023Inspiration Friday: 2023 AMD World Bike Building Championships • Total Motorcycle
Aug 14, 2023AXERA AX620A 4K AI SoC delivers up to 14.4 TOPS for computer vision applications
May 16, 2023AXERA AX620A is a high-performance, low-power AI SoC with a quad-core Arm Cortex-A7 processor and a 14.4TOPs @ INT4 or 3.6TOPs @ INT8’s NPU that is slightly inferior to the Amlogic A311D, and mainly used for AI vision applications.
With high computing power and built-in image processing capabilities, the AX620A can support a wide range of AI workloads. It also offers low power consumption with low standby power and fast wake-up, so the chip can be integrated into battery-powered products.
AXERA AX620A specifications:
AXERA has developed the AI ISP image processing technology in-house with the combination of the ISP and NPU improving the performance and features compared to more traditional ISPs. The AX620A’s AI ISP is said to achieve “excellent image quality” in low-light environments and does so that a low power consumption of around 1 Watt, and it may be integrated into other Edge AI chips. Let’s wait and see.
We found what appears to be the official AX620A development board in an article posted on Baidu which comes with 2GB RAM and 8GB eMMC flash and offers a range of ports including Gigabit Ethernet, two MIPI CSI interfaces, one MIPI DSI display interface up to 4Kp30, two relays, two RS485 interfaces, and more. What’s not shown in that article is software support, but AXERA’s GitHub, has some code with Pipeline examples of AX620A’s ISP, image, NPU, and other modules, as well as the sample code for computer vision. The AX620A SDK is based on Debian 11 Linux.
You may also visit the product page to get more information, but there are only a few details beyond the hardware specs.
Li Zexi became a part-time writer for CNX Software in 2022 and works as a software engineer in an embedded software company in China. He’s been passionate about embedded development and the Internet of Things for many years, with extensive industry experience, and a new and unique vision!
Support CNX Software! Donate via cryptocurrencies or become a Patron on Patreon
Related posts:
NPU – 4 TOPS @ INT4, 3.6 TOPS @ INT8 with support for Imagenet, AlexNet, VGG, ResNet, GoogLeNet, Faster R-CNN, SSD, FPN, Yolo V3, and other neural networks.