News for hackers. Stuff that matters.

A comment on YOLOv6: Redefine state-of-the-art for object detection.

Re:YOLOv6: Redefine state-of-the-art for object detection by notme1234 on Wednesday June 29, 2022 @05:13PM (#31925787)
What is the best object detection that is able to run on Raspberry PI 2? I checked tiny Yolo which requires less resources but it is much less accurate than the regular Yolo. Looking for the best accuracy with the least CPU/RAM requirements. Mainly needs this to tell if an image contain a person (still images from a network camera, which publishes an image upon motion detecion).
[ Parent | on HN ]
Re:YOLOv6: Redefine state-of-the-art for object detection by sdlion on Thursday June 30, 2022 @03:09AM (#31929519)
It would seem that it depends on the architecture you will be using. Whether is a ARM, GPU, mobile GPU processor, etc. This is comment from the author of YOLOv4 mentions that NanoDet is more suitable for ARM-CPU's https://twitter.com/alexeyab84/status/1436377831974506496
[ Parent | on HN ]