Simple video page scanner that applies an homography in all video frames of the video and joins them.
It receives a video containing a paper (example) and outputs a video only with the paper changing along with the original video.
This project was made for the Computer Vision course.
Install dependecies
pip3 install -r requirements.txt
Run
python3 main.py -v VIDEO -o OUTPUT -t TEMPLATE
VIDEO
- Video input file containing the paper (default: video.mp4)OUTPUT
- Name of the output video file (default: output.mp4)TEMPLATE
(optional) - Image file containing the wanted dimensions of output paper in the video (default dimensions: 1275x1650)
You can run the program with the example video file
video.mp4
- Expected output
output.mp4
Input | Output |
---|---|
) |
- Original project statement: statement.pdf