rife's Introduction

RIFE: Real-Time Intermediate Flow Estimation for Video Frame Interpolation

Creates sequence of interpolated frames between given input images

Run

python interpolate.py --input samples/ --output frames/ --buffer 25 --multi 25

interpolating 4 images
image samples/image000.jpg ssim 0.99 buffer 25 frames
image samples/image001.jpg ssim 0.54 create 69 frames
image samples/image002.jpg ssim 0.45 create 69 frames
image samples/image003.jpg ssim 0.55 create 69 frames
image samples/image003.jpg ssim 0.99 buffer 25 frames
frames 259 time 4.24

Reads input images from samples/ and writes output images to frames/
Number of generated frames will be 70x input frames
Start and end will be buffered/padded with 25 frames

ffmpeg -hide_banner -loglevel warning -hwaccel auto -y -framerate 30 -i "frames/%6d.jpg" -r 30 -vcodec libx264 -preset medium -crf 23 -vf minterpolate=mi_mode=blend,fifo -movflags +faststart samples/video.mp4

Creates a video file from interpolated frames

Options

./interpolate.py --help

--model MODEL    path to model
--input INPUT    input directory containing images
--output OUTPUT  output directory for interpolated images
--scale SCALE    scale factor for interpolated images
--multi MULTI    number of frames to interpolate between two input images
--buffer BUFFER  number of frames to buffer on scene change
--change CHANGE  scene change threshold (lower is more sensitive
--fp16           use float16 precision instead of float32

Example

Both examples are created using SD.Next

Using AnimateDiff extension

rife.mp4

Video: 2.5sec at 25fps using 16 input images

Using Seed Travel extension

video.mp4

Video: 9sec at 30fps using 10 input images

Credits

rife's People

Contributors

Stargazers

Watchers

rife's Issues

Question regarding model?

I don't really have an "issue" per-se I just had a general question if you wouldn't mind answering, the girl in the pictures that you've uploaded and her body looks quite realistic, would you mind sharing what stable diffusion model, sampling method & sampling steps and other settings that you used to create it? I would really really appreciate it.

Thanks!