5.8 KiB
RIFE Video Frame Interpolation v1.4
arXiv | Project Page | Reddit | YouTube
11.30 News: We have updated the v1.4 model to greatly reduce the patch artifacts when the camera moves vigorously. Please check our update log.
11.22 News: We notice a new windows app is trying to integrate RIFE, we hope everyone to try and help them improve. You can download Flowframes for free.
There is a tutorial of RIFE on Youtube.
You can easily use colaboratory to have a try and generate the our youtube demo.
Our model can run 30+FPS for 2X 720p interpolation on a 2080Ti GPU. Currently, our method supports 2X,4X interpolation for 1080p video, and multi-frame interpolation between a pair of images. Everyone is welcome to use our alpha version and make suggestions!
16X interpolation results from two input images:
Usage
Installation
git clone git@github.com:hzwer/arXiv2020-RIFE.git
cd arXiv2020-RIFE
pip install -r requirements.txt
- Download the pretrained models from here. We are optimizing the visual effects and will support animation in the future.
(我们也提供了百度网盘链接:https://pan.baidu.com/s/1TniBk2Ld2O6XXgr-AY9DMQ 密码:ub97，把压缩包解开后放在 train_log/*.pkl)
- Unzip and move the pretrained parameters to train_log/*.pkl
Run
Video Frame Interpolation
You can use our demo video or your own video.
python3 inference_video.py --exp=1 --video=video.mp4
(generate video_2X_xxfps.mp4)
python3 inference_video.py --exp=2 --video=video.mp4
(for 4X interpolation)
python3 inference_video.py --exp=2 --video=video.mp4 --fps=60
(add slomo effect, the audio will be removed)
python3 inference_video.py --video=video.mp4 --montage --png
(if you want to montage the origin video, and save the png format output)
The warning info, 'Warning: Your video has *** static frames, it may change the duration of the generated video.' means that your video has changed the frame rate by adding static frames, it is common if you have processed 25FPS video to 30FPS.
Image Interpolation
python3 inference_img.py --img img0.png img1.png --exp=4
(2^4=16X interpolation results) After that, you can use pngs to generate mp4:
ffmpeg -r 10 -f image2 -i output/img%d.png -s 448x256 -c:v libx264 -pix_fmt yuv420p output/slomo.mp4 -q:v 0 -q:a 0
You can also use pngs to generate gif:
ffmpeg -r 10 -f image2 -i output/img%d.png -s 448x256 -vf "split[s0][s1];[s0]palettegen=stats_mode=single[p];[s1][p]paletteuse=new=1" output/slomo.gif
Evaluation
First, you should download RIFE model reported by our paper.
We will release our training and benchmark validation code soon.
Vimeo90K: Download Vimeo90K dataset at ./vimeo_interp_test
MiddleBury: Download MiddleBury OTHER dataset at ./other-data and ./other-gt-interp
python3 benchmark/Vimeo90K.py
# (Final result: "Avg PSNR: 35.695 SSIM: 0.9788")
python3 benchmark/MiddelBury_Other.py
# (Final result: "2.058")
Training
Because Vimeo90K dataset and the corresponding optical flow labels are too large, we cannot provide a complete dataset download link. We provide you with a subset containing 100 samples for testing the pipeline. Please unzip it at ./dataset
For origin images, you can download them from Vimeo90K dataset.
For generating optical flow labels, our paper use pytorch-liteflownet. We also recommend RAFT because it's easier to configure. We recommend generating optical flow labels on 2X size images for higher quality. You can also generate labels during training, or finetune the optical flow network on the training set. The final impact of the above operations on Vimeo90K PSNR is expected to be within 0.3.
We use 16 CPUs, 4 GPUs and 20G memory for training:
python3 -m torch.distributed.launch --nproc_per_node=4 train.py
Citation
@article{huang2020rife,
title={RIFE: Real-Time Intermediate Flow Estimation for Video Frame Interpolation},
author={Huang, Zhewei and Zhang, Tianyuan and Heng, Wen and Shi, Boxin and Zhou, Shuchang},
journal={arXiv preprint arXiv:2011.06294},
year={2020}
}
Reference
Optical Flow: ARFlow pytorch-liteflownet RAFT pytorch-PWCNet
Video Interpolation: DAIN CAIN TOflow MEMC-Net SoftSplat SepConv BMBC