Skip to content
Using YOLO9000, classify objects in a video
Python
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
.gitignore
README.md
strain.py

README.md

Video Object Recognition

Using YOLO9000, classify objects in a video (made to support a wider range of computers than YOLO-9000 supports)

Dependencies

Usage

  1. Run ffmpeg -i video.mp4 "frames/out-%08d.jpg" on your desired video
  2. Put the frames folder into YOLO-9000's folder
  3. Put strain.py into darknet/ and run the script
  4. Concatenate the frames back into a video with ffmpeg -framerate 25 -i out_%08d.jpg output.mp4 (framerate will need adjusting)

Known issues

  • The script takes a long time to run. This is because the nueral network is running on the CPU for compatability reasons. Machine learning is much faster on a GPU but very few GPUs fully support NN/ML. To remedy, run overnight.

Example

I collected clips of Sydney and fed it through the classifier and paired it with music to produce this result.

You can’t perform that action at this time.