High Performance GMFSS with RIFE and GAN for Video Frame Interpolation
2023-04-03: We now provide GMFSS_Fortuna as a factual basis for training in GMFSS. Please use it. This item will not be updated!
python3 inference_video.py --img=demo/ --scale=1.0 --multi=2
2023.02.10,The training code has been re-released, although nothing has actually changed. The official training log and results can be found in Google Drive. Please note that this result is only used to verify the training process, for inference please still use the version released above. Finally I deleted the RIFE_Fix_GAN_Loss_output.py, which only increases the misunderstanding
2022.11.27, the training code has been made public. This code is run on a single V100, so there is no DistributedDataParallel related code. Please refer to RIFE train. The required training set is ATD-12K, which can be obtained from the official library. We also used a private data set, which unfortunately cannot be made public for the time being.In the process of organizing the code in the future, I found an error in the numerical processing of GAN Loss. I provided a replacement file, but the overall situation still maintained my original processing.
This project is supported by SVFI Development Team