AAEON's new ARM based solution.
summary
Compact ARM edge box called Roxer is based on the Rainy board and using a Rockchip 339 with an on‑chip NPU to run vision analytics. It demonstrates person and chair recognition at about ~30 fps person/chair detection while drawing ~5 W power and stabilizing around 26 °C operation. Shipping form factor is a small fanless box aimed at local inference rather than cloud offload, with emphasis on low power and real‑time performance.
AI-written summary · about
related talks
16mRVEdge-Vision: A Fully Open,Ultra-Efficient On-Device AI Platform for Smart Eyewear
14mEmbedded Executive LIVE
12mCost-effective Hardware Accelerator Recommendation for Edge Computing
28mML on the Edge for Industry 4.0 with Arm Ethos N-78 Neural Processing Unit
06mEmbedded Executive LIVE
02mKen Briodagh interviews Irene Yang
16mIntroducing Coral Dev Board Micro
24mInnovation and Solutions for Embedded and Edge AI