#trialrun from a-z for one video

[brialorelle]

• write the scripts with sanity checks for frame extraction and frame timing.

1. Extract frames
2. Count frames
3. Show frames
4. Rename frames
5. Save frames
6. Move frames
7. Save .csv with all info

• Grab one or two videos (one alice, one sam) of reasonable length and run mtcnn on frames
• Make small gold set for face detection on these videos
• Evaluate MTCNN performance