Initial commit

README.md

@@ -0,0 +1,35 @@
+# Video Inference
+
+This repository is an inference repo similar to that of the ESRGAN inference repository, but for various video machine learning models.
+
+## Currently supported architectures
+
+- SOFVSR (@Victorca25's BasicSR Version)
+
+## Additional features
+
+- Automatic scale, number of frames, and number of channels detection
+- Automatic beginning and end frame padding so frames 1 and -1 get included in output
+
+## Using this repo
+
+- Place exported video frames in the `input` folder
+- Place model in the `models` folder
+- `python run.py ./models/video_model.pth`
+
+## Extra flags
+
+- `--input`: Specifies input directory
+- `--output`: Specifies output directory
+- `--denoise`: Denoises the chroma layer
+- `--chop_forward`: Splits tensors to avoid out-of-memory errors
+
+## Planned architecture support
+
+- EDVR
+- RRN
+- RIFE
+
+## Planned additional features
+
+- Direct video input/output via ffmpeg