video2vr

What does it do?

Easily convert videos into side-by-side VR videos.

How does it work?

The program estimates depth maps for each frame using AI.
Using these depth maps it shifts the pixels for the left and right eye independendly.
Then we place those 2 frames side-by-side and get a VR-video.

How to start the program?

git clone https://github.com/xXChampionsXx/video2vr
cd video2vr
pip3 install -r requirements.txt
python video2vr.py

GPU support / CUDA support

Install pytorch with cuda support following the instructions on here: https://pytorch.org/get-started/locally/#start-locally