Avant-Gardiste
commented
2 years ago Yolov5 is too big to run it on a mobile device in reasonable time. I would suggest to train it on Yolov5n (nano) instead of Yolov5s (small).
Open blackCmd opened 2 years ago
Avant-Gardiste
commented
2 years ago Yolov5 is too big to run it on a mobile device in reasonable time. I would suggest to train it on Yolov5n (nano) instead of Yolov5s (small).
Is there a way to increase the detection speed during the live object detection?
It is so slower than tensorflow lite object detection.