search

zzh-tech / ESTRNN

[ECCV2020 Spotlight] Efficient Spatio-Temporal Recurrent Neural Network for Video Deblurring
MIT License
314 stars 39 forks source link

Using pre-trained model to inference , the results is strange #36

Closed wangwyForNewTime closed 2 years ago

wangwyForNewTime commented 2 years ago

I choose 15 pictures from BSD dataset to test I use the comman： python inference.py --src ./blur/ --dst ./results/ --ckpt ./checkpoints/ESTRNN_C80B15_BSD_2ms16ms.tar

the result is so strange 12 13

I prepare a video to test Use the comman: python inference.py --src 1.mp4 --dst ./results/2/ --ckpt ./checkpoints/ESTRNN_C80B15_BSD_2ms16ms.tar there is an error: 14

Anything I did wrong?

zzh-tech commented 2 years ago

Please share the test data with me. I'll check it for you later.

wangwyForNewTime commented 2 years ago

Please share the test data with me. I'll check it for you later.

thank you BaiduNetdiskExtractCode：1024 There is my test data,I tried it on both windows and linux ,the results are same

zzh-tech commented 2 years ago

I tested the case with the BSD image sequence as input. There is no problem. The issue may be on your side.

Regarding the video, it seems that the actual number of frames does not match the number read by cv2. Anyway, I've updated the code that extracts frames from the video to avoid this.

wangwyForNewTime commented 2 years ago

I tested the case with the BSD image sequence as input. There is no problem. The issue may be on your side.

Regarding the video, it seems that the actual number of frames does not match the number read by cv2. Anyway, I've updated the code that extracts frames from the video to avoid this.

Thank you for your reply There is no error reminder when test BSD images ，I think the strange result maybe because of the different version of environment library version Accroding to ReadMe: 11 I use Annaconda to create a new environment,there is conda list: 12

It could be the version，can you give me the specific version which you are using?

zzh-tech commented 2 years ago

The list of dependencies of my current env (generated by pipreqs) is as follows for your reference:

lmdb==1.3.0 numpy==1.21.5 opencv_python==4.5.3.56 Pillow==9.2.0 rawpy==0.17.1 scikit_image==0.19.2 thop==0.1.0.post2207010342 torch==1.11.0 torchvision==0.12.0 tqdm==4.64.0

wangwyForNewTime commented 2 years ago

The list of dependencies of my current env (generated by pipreqs) is as follows for your reference:

lmdb==1.3.0 numpy==1.21.5 opencv_python==4.5.3.56 Pillow==9.2.0 rawpy==0.17.1 scikit_image==0.19.2 thop==0.1.0.post2207010342 torch==1.11.0 torchvision==0.12.0 tqdm==4.64.0

Thank you,the problem is solved. This is really version problem.Now coding runs smoothly In addition to the above, also need python 3.7