pol22 / videoupsampler

Video Upsampler based on Optical Flow from previous and next frames

VideoUpsampler

MP4 video upsampler

Video

Paired video with lower resolution was used for training

Using ffmpeg can be generated downsampled video

Example from 1080p with x2 downsampled scale:

ffmpeg -i input/mp4/video -s 960x540 -crf 30 -vsync vfr output/mp4/video

Model

Model based on UNet architecture (more in tf_model.py)

Trained on MP4 encoded videos with lower resolution

Pretrained x2 TF model

Pretrained x3 TF model

Usage

foo@bar:~/VideoUpsampler$ python upsampler.py --file path/to/mp4/file --model path/to/model 

New video will be saved near original file with _upsampled suffix

Also suffix _audio will be added for video with copied audio

YouTube example