:zap: Based on yolo's ultra-lightweight universal target detection algorithm, the calculation amount is only 250mflops, the ncnn model size is only 666kb, the Raspberry Pi 3b can run up to 15fps+, and the mobile terminal can run up to 178fps+
I'm using OpenCV.
When I use an input image of 1280x720 or similar aspect ratio, objects are detected correctly regardless of their position.
But if I use one of 640x480 or similar aspect ratio, they are only detected from a certain Y coordinate down the image, something like the bottom 2/3
I'm using OpenCV. When I use an input image of 1280x720 or similar aspect ratio, objects are detected correctly regardless of their position. But if I use one of 640x480 or similar aspect ratio, they are only detected from a certain Y coordinate down the image, something like the bottom 2/3