Resnet output doesn't match the FPN_FEAT_STRIDES

jwyang / fpn.pytorch

Pytorch implementation of Feature Pyramid Network (FPN) for Object Detection

MIT License

955 stars 224 forks source link

Resnet output doesn't match the FPN_FEAT_STRIDES #12

Closed destinyzs closed 3 years ago

destinyzs commented 6 years ago

The default cnn network is res101. However, on the fpn.py line 184, the size of p2, p3, p4, p5, p6 are 1/4,1/8,1/16,1/16,1/32 of the input image's size, so the FPN_FEAT_STRIDES should be [4,8,16,16,32]. But on the trainval_net.py line 173, default FPN_FEAT_STRIDES is [4, 8, 16, 32, 64], the config and code don't match. I not sure whether the result will be affected, but i think you can modify it. @jwyang

cjxnew commented 3 years ago

Good suggestion brother, but I guess the author is not deal with this project any more......

destinyzs commented 3 years ago

yeah... so i will close this issue