BeagleBoard has launched a new single-board computer called the BeagleV-Ahead. It’s the same shape as the company’s BeagleBone Black and it’s compatible with some accessories designed for that board.
But instead of an ARM-based processor, the new BeagleV-Ahead is powered by a quad-core RISC-V processor. The new board is available now for around $149.
At the heart of the new board is an Alibaba T-Head TH1520 chip which features four 2 GHz Xuantie C910 CPU cores, Imagination BXM-4-64 integrated graphics and a neural processing unit with up to 4 TOPS of AI performance.
Yeah, if your objects within the 4k frame are pretty small, you can definitely tile over the whole image with 192x192 windows. With YOLOv8n (the smallest model size), I could run inference on 192x192 images in 15ms per inference on average, and about 25ms per inference for 192x192 images on YOLOv8s (the second-smallest model size). We’ve been getting pretty good results in terms of accuracy and so forth with both n and s.
Biggest benefit of YOLOv8, imo, is it has a really good software package that makes it very easy to finetune, export to the edge TPU (or other inference engines), and run inference. Feel free to message me if you run into any issues with it!
Yea would probably have to do a several passes at different resolutions. Looking for liquid pooling so doesn’t have to be fast, just has to be right