Grazper's Smart city solution
summary
Grazper ports an AI Inference Engine on FPGA to AAEON hardware and demonstrates it running on UP AI Plus (Intel Cyclone 10) and UP Core Plus boards. The build implements tiny YOLO at about 20 frames per second on that board and targets around 35 fps with optimizations. The implementation supports multiple FPGA families and AAEON platforms, showing a working inference pipeline rather than a simulation and concrete throughput numbers for edge smart city vision.
AI-written summary · about
related talks
10mRealizing the Power of AI at the Edge
19mMission-Critical AI in Space and Sky: SWaP-Constrained Intelligence Wi...
50mAI acceleration at the edge
36mApplied AI on the edge
20mScaling Edge AI with Elixir and Nerves
22mBuilding the AI City: Edge-to-Cloud AI Inference for Smart Cities
13mDiscover What's Possible With Fast AI Inference At Groq Speed
13mAccelerating AI algorithms