I made a mistake with my newly bought compact camera, before a trip to Japan. Instead of taking single pictures, I ended up with a few short "MP4" video files (1 second, 2-3 frames). I could "easily enough" extract individual frames with ffmpeg, using help from:

I am unsatisfied with the quality of some frames, and as there is very little motion, I would like to restore or recreate a superresolution image from those two-three frames only, using the motion vector.

I would appreciate suggestions, software experience, etc. For readers interested in trying algorithms, I provided in the 22 MB file StackExchange-AVI2JPG.zip two examples with media information (obtained with MediaInfo) and camera reference (to avoid unwanted advertizing). One leaving Osaka, with a relatively nicely flat image:

Leaving Osaka in the bus

the other from Nara, Todai-Ji:

Nara, Todai-Ji

where the wire fence is sharper, hence the individual frame quality is poorer. I enjoyed my trip, and would be glad to share crisper images of the first day (a view on the Fuji in the early for the finest eyes).


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.