-
Notifications
You must be signed in to change notification settings - Fork 86
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
The video file '/assets/data/EMOCA_test_example_data/videos/82-25-854x480_affwild2.mp4' is corrupted. Skipping it. #75
Comments
I have the same issue, did you fix it? |
FYI: I fixed the problem. |
Yeah, I have also fixed the issue, in my case I think there was the wrong path to the video. |
I faced the same issue. But after running on my own video it finally works |
I faced the same issue,I changed my own video and still reported this error,Do you know how to fix this |
I don't know why I am getting that error, The path is correct and video is also fine.
The video file '/assets/data/EMOCA_test_example_data/videos/82-25-854x480_affwild2.mp4' is corrupted. Skipping it.
Processing dataset
0%| | 0/1 [00:00<?, ?it/s]/assets/data/EMOCA_test_example_data/videos/82-25-854x480_affwild2.mp4
The video file '/assets/data/EMOCA_test_example_data/videos/82-25-854x480_affwild2.mp4' is corrupted. Skipping it.
100%|███████████████████████████████████████████████████████████████████████████████████████| 1/1 [00:00<00:00, 55.22it/s]
Traceback (most recent call last):
File "gdl_apps/EMOCA/demos/test_emoca_on_video.py", line 172, in
main()
File "gdl_apps/EMOCA/demos/test_emoca_on_video.py", line 167, in main
reconstruct_video(args)
File "gdl_apps/EMOCA/demos/test_emoca_on_video.py", line 62, in reconstruct_video
dm.prepare_data()
File "/home/abbas/anaconda3/envs/emoo36/lib/python3.8/site-packages/pytorch_lightning/core/datamodule.py", line 428, in wrapped_fn
fn(*args, **kwargs)
File "/home/abbas/emoca/gdl/datasets/FaceVideoDataModule.py", line 3192, in prepare_data
self._gather_data(exist_ok=True)
File "/home/abbas/emoca/gdl/datasets/FaceVideoDataModule.py", line 3214, in _gather_data
self._gather_video_metadata()
File "/home/abbas/emoca/gdl/datasets/FaceVideoDataModule.py", line 1570, in _gather_video_metadata
del self.annotation_list[vi]
IndexError: list assignment index out of range
The text was updated successfully, but these errors were encountered: