ObjectDetection-OneStageDet icon indicating copy to clipboard operation
ObjectDetection-OneStageDet copied to clipboard

why mobilev1/v2 only use middle features to predict?

Open wangxinzhe0617 opened this issue 6 years ago • 3 comments

why mobilev1/v2 only use middle features to predict? The result is better?

wangxinzhe0617 avatar Apr 16 '19 08:04 wangxinzhe0617

I just make a demo, you can modify it.

mileistone avatar May 05 '19 09:05 mileistone

ok,Thank you for your reply.

wangxinzhe0617 avatar May 05 '19 09:05 wangxinzhe0617

I also want to know how to modify it using all features...

hilmandayo avatar Sep 10 '19 02:09 hilmandayo