Closed jblee4sphinx closed 4 years ago
It seems to me that "../docs/examples/.mp4" is not an actual path to a video. What is the full command you used?
I removed the video filename prior to pasting. Here's the full command being run. The path is relative, similarly to how you did the shinzo and conan examples. I copied my source and target videos into the same directory with examples.
(base) root@ip-10-0-10-227:/data/dev/projects/fsgan/inference# python swap.py ../docs/examples/source_video.mp4 -t ../docs/examples/target_video.mp4 -o . --finetune --finetune_save --seg_remove_mouth
Try re-encoding both videos with a standard encoding:
ffmpeg -i <input video> -r 25 <output video>
I finished re-encoding but now retrying the swap.py gives me an Assertion Error with CUDA. File "/data/dev/projects/face_detection_dsfd/face_detector.py", line 19, in init assert torch.cuda.is_available(), 'CUDA must be available!' AssertionError: CUDA must be available!
Is there a fix for this?
Did you install the CUDA toolkit and NVIDIA driver?
Yes. I followed every command step-by-step from https://github.com/YuvalNirkin/fsgan/wiki/Ubuntu-Installation-Guide. I will just rebuild and start over. Thanks for the help.
I had the same issue and can confirm that re-encoding both videos fixed the problem.
I'm having this issue and re-encoding is not fixing the problem. Is there another workaround?
Hello, I get this error at the end of processing swap.py. Is there a fix for this? Thank you!
Process FaceSwappingRenderer-1: Traceback (most recent call last): File "/root/anaconda3/lib/python3.7/multiprocessing/process.py", line 297, in _bootstrap self.run() File "/data/dev/projects/fsgan/utils/video_renderer.py", line 159, in run assert frame_bgr is not None, f'Failed to read frame {i} from input video: "{self._in_vid_path}"' AssertionError: Failed to read frame 1253 from input video: "../docs/examples/