search

jfzhang95 / pytorch-video-recognition

PyTorch implemented C3D, R3D, R2Plus1D models for video activity recognition.
MIT License
1.16k stars 250 forks source link

Wrong construction of the first conv layer in R2Plus1D_model.py #62

Open Taylor-X76 opened 3 years ago

Taylor-X76 commented 3 years ago

In R2Plus1D_model.py, line 200： https://github.com/jfzhang95/pytorch-video-recognition/blob/ca37de9f69a961f22a821c157e9ccf47a601904d/network/R2Plus1D_model.py#L200

It's actually a convolution of 3 7 7 with padding=(1, 3, 3), not 1 7 7！