Skip to content

Demetter/TSCUNet_Trainer

 
 

Repository files navigation

Python Train_TSCUNet.py

Python Train_TSCUNet.py --Checkpoint Dataset/Models/Checkpoint_Epoch.py


VidToFrames.py will extract frames from video and rename them correctly. Frames must be in multiples of 5 and sequential.
If you don't use the script they must be named as shown below. VAL_HR/LR must have sequential frames in multiples of 5.

About

Temporal Upscaling via Swin-Conv-UNet

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 100.0%