Loss changes little (~4.5) and Prec@1 keeps extremely low (~2%) when training with flow images about two-stream-pytorch HOT 4 CLOSED

Hi, thanks for bringing up the issue. Actually I think it is because the pre-processing of optical flow images.

The default preprocessing of PyTorch for natural color images is:

clip_mean = [0.485, 0.456, 0.406] * args.new_length
clip_std = [0.229, 0.224, 0.225] * args.new_length

But for optical flow, I haven't found an effective way to normalize it in order to use the pre-trained models. Currently, i use 0.5 because most people use 0.5 for gray scale images. But it doesn't work.

I will debug into it later. Or you can make progress on this issue. Thanks.

from two-stream-pytorch.

@bryanyzhu Oh, I just saw your reply. Sorry about this.
I tried different mean and std values, but it didn't help. Do you make any progress?
Can you share your results (spatial and temporal) of this code on UCF101?

from two-stream-pytorch.

@yanjunzhu Hi, i just updated the codebase and report the accuracy on both spatial and temporal model on UCF101. Hopefully you can train your flow models now. But as I mentioned in README, my flow model's accuracy is still 5% lower than it reported in the paper. So maybe there still exist some subtle problems. Welcome any comments and pull request. Thank you.

from two-stream-pytorch.

@bryanyzhu Thanks. I will have a try later.

from two-stream-pytorch.