Encoding error

[myfoxx]

python blur_faces.py media/friends.mp4 --mode all --censor-type pixelation
media/friends.mp4
mode='all', model='hog', censor_type='pixelation', count=1, in_face_file=None
width=1280, height=720, length=157, fps=23.976, codec='avc1'
[ERROR:0@0.024] global /io/opencv/modules/videoio/src/cap_ffmpeg_impl.hpp (2927) open Could not find encoder for codec_id=27, error: Encoder not found
[ERROR:0@0.024] global /io/opencv/modules/videoio/src/cap_ffmpeg_impl.hpp (3002) open VIDEOIO/FFMPEG: Failed to initialize VideoWriter
99%|████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████▋ | 157/158 [01:13<00:00, 2.13it/s]
ffmpeg version N-114385-g4743c9e87a Copyright (c) 2000-2024 the FFmpeg developers
built with gcc 9 (Ubuntu 9.4.0-1ubuntu1~20.04.2)
configuration: --enable-cuda-nvcc --enable-cuvid --enable-nvenc --enable-nonfree --enable-libnpp --extra-cflags=-I/usr/local/cuda/include --extra-ldflags=-L/usr/local/cuda/lib64
libavutil 59. 7.100 / 59. 7.100
libavcodec 61. 2.100 / 61. 2.100
libavformat 61. 0.100 / 61. 0.100
libavdevice 61. 0.100 / 61. 0.100
libavfilter 10. 0.100 / 10. 0.100
libswscale 8. 0.100 / 8. 0.100
libswresample 5. 0.100 / 5. 0.100
[mov,mp4,m4a,3gp,3g2,mj2 @ 0x5609d1e68000] Format mov,mp4,m4a,3gp,3g2,mj2 detected only with low score of 1, misdetection possible!
[mov,mp4,m4a,3gp,3g2,mj2 @ 0x5609d1e68000] moov atom not found
[in#0 @ 0x5609d1e67cc0] Error opening input: Invalid data found when processing input
Error opening input file /tmp/tmpf9ffrai0.mp4.
Error opening input files: Invalid data found when processing input
Traceback (most recent call last):
File "blur_faces.py", line 143, in
blurfaces()
File "/opt/BlurryFaces/src/env/lib/python3.8/site-packages/click/core.py", line 1130, in call
return self.main(*args, **kwargs)
File "/opt/BlurryFaces/src/env/lib/python3.8/site-packages/click/core.py", line 1055, in main
rv = self.invoke(ctx)
File "/opt/BlurryFaces/src/env/lib/python3.8/site-packages/click/core.py", line 1404, in invoke
return ctx.invoke(self.callback, **ctx.params)
File "/opt/BlurryFaces/src/env/lib/python3.8/site-packages/click/core.py", line 760, in invoke
return __callback(*args, **kwargs)
File "blur_faces.py", line 137, in blurfaces
ffmpeg.run(stream)
File "/opt/BlurryFaces/src/env/lib/python3.8/site-packages/ffmpeg/_run.py", line 325, in run
raise Error('ffmpeg', out, err)
ffmpeg._run.Error: ffmpeg error (see stderr output for detail)

My ffmpeg

ffmpeg -hwaccels
ffmpeg version N-114385-g4743c9e87a Copyright (c) 2000-2024 the FFmpeg developers
built with gcc 9 (Ubuntu 9.4.0-1ubuntu1~20.04.2)
configuration: --enable-cuda-nvcc --enable-cuvid --enable-nvenc --enable-nonfree --enable-libnpp --extra-cflags=-I/usr/local/cuda/include --extra-ldflags=-L/usr/local/cuda/lib64
libavutil 59. 7.100 / 59. 7.100
libavcodec 61. 2.100 / 61. 2.100
libavformat 61. 0.100 / 61. 0.100
libavdevice 61. 0.100 / 61. 0.100
libavfilter 10. 0.100 / 10. 0.100
libswscale 8. 0.100 / 8. 0.100
libswresample 5. 0.100 / 5. 0.100
Hardware acceleration methods:
vdpau
cuda
drm

[deyang2000]

It appears that you are using Python version 3.8, as indicated by the path "/opt/BlurryFaces/src/env/lib/python3.8/site-packages/ffmpeg/_run.py". I recommend upgrading to Python version 3.10.6 to enhance compatibility and improve performance. However, please avoid upgrading to version 3.11, as it might cause compatibility issues with some of the packages you're using.

For optimizing your code, consider replacing the existing segment responsible for setting up video writing with the following lines:

 fourcc = cv2.VideoWriter_fourcc(*'mp4v') 
 with tempfile.NamedTemporaryFile(suffix=file_extension) as out_video_file:
        video_out = cv2.VideoWriter(
            out_video_file.name, fourcc, fps, (width, height))

Everything functions correctly on my end with the current setup, but the processing speed is notably slow. For instance, it took 27 minutes to run the demo. This indicates there might be room for further optimization, especially in terms of how frames are processed and written concurrently.

Consider implementing techniques such as multi-threading or utilizing more efficient libraries or algorithms for video processing, which might significantly reduce the overall processing time. Additionally, checking for hardware acceleration support (like CUDA for dlib in face recognition) could be beneficial.

[raviksharma]

I changed the library, and it is much faster on my cpu. I am closing this bug since it is not relevant any more.