Hi all, I have a specific task I want to automate - rotating a video file through 90 degrees.
I've used the PIL library quite a bit to perform batch processing on images, and would like to do similar on video. Can anyone see a problem with the following: 1) Use pymedia to convert the video into a sequence of stills (e.g. http://pymedia.org/tut/src/dump_video.py.html) 2) Use PIL to process each still 3) Use pymedia to re-pack the still into video format. In particular, does anyone know whether the data obtained from decoding the video frame as in the following snippet from http://pymedia.org/tut/src/dump_video.py.html: dd= d.convert( fmt ) img= pygame.image.fromstring( dd.data, dd.size, "RGB" ) can be directly loaded into a PIL Image object (and back again?)? -- http://mail.python.org/mailman/listinfo/python-list
