PySceneDetect
PySceneDetect copied to clipboard
can`t save_images in scenedetect
I have a scene list
[(00:18:17.467 [frame=32924, fps=30.000], 00:19:17.467 [frame=34724, fps=30.000]), (00:47:14.200 [frame=85026, fps=30.000], 00:48:14.200 [frame=86826, fps=30.000]), (00:48:14.200 [frame=86826, fps=30.000], 00:49:14.200 [frame=88626, fps=30.000]), (00:49:14.200 [frame=88626, fps=30.000], 00:50:14.200 [frame=90426, fps=30.000]), (00:50:14.200 [frame=90426, fps=30.000], 00:51:14.200 [frame=92226, fps=30.000]), (00:51:43.333 [frame=93100, fps=30.000], 00:52:43.333 [frame=94900, fps=30.000]), (00:52:43.333 [frame=94900, fps=30.000], 00:52:48.533 [frame=95056, fps=30.000]), (00:59:41.667 [frame=107450, fps=30.000], 01:00:36.467 [frame=109094, fps=30.000]), (01:00:36.467 [frame=109094, fps=30.000], 01:01:36.467 [frame=110894, fps=30.000]), (01:01:36.467 [frame=110894, fps=30.000], 01:02:36.467 [frame=112694, fps=30.000]), (01:02:52.933 [frame=113188, fps=30.000], 01:03:52.933 [frame=114988, fps=30.000]), (01:04:17.067 [frame=115712, fps=30.000], 01:05:17.067 [frame=117512, fps=30.000]), (01:05:17.067 [frame=117512, fps=30.000], 01:06:17.067 [frame=119312, fps=30.000])]
I try to save some screenshots from scenes to file but i get error
help_vid = open_video(clip_path)
write_scene_list_html(output_html_filename="%s_before_concat_list.html" % clip_path, scene_list=output_scene)
save_images(video=help_vid,scene_list=output_scene, num_images = 5, image_name_template="%s_help_$SCENE_NUMBER-$IMAGE_NUMBER.mp4"%clip_path , width=200, height=200)
Could you please help, can
t understand what ive done wrong
If this is just a warning how to ignore it and continue my script?
This error is indicating that decoding the desired frame of the video was not successful. The save_images function seeks to the desired frame to pull as an image here and then calls the read function to decode the desired frame in the next line. It's not clear which backend you are using, but you can see the docstring for the read function when using the OpenCV backend (the default backend) that it is supposed to return a decoded numpy array and a boolean. In cases in which the decoding is not successful, then just a boolean is returned. This would cause the error that you see here.
Does this happen with other videos, or is it specific to one video?
yes, this happens to other videos as well
is there any way to skip corrupted frame during image_saving (working of function save_images)? in order to save all other "good" frames
Does the issue still happen once you re-encode (or re-mux) the input video? The OpenCV backend tries to recover from this gracefully, so I assume seeking isn't happening correctly for this video.
As in #345, do the other videos this happens with also have trouble seeking?
Could you please help me with this error FrameTimecode instances require equal framerate for addition
Could anyone give me the solution? I faced the same error
@KerolosAtef Can you try the PyAV backend (or MoviePy)? Otherwise you may have to re-encode / re-mux the source material first. Can you provide a sample of a clip that causes this error?
If anyone can provide a reproduction of the issue, that would be very helpful.
@Vsevosemnog can you provide some more context in a new issue/discussion? Thanks!
Closing this issue due to lack of context, feel free to re-open or file a new one if this still occurs.