uglyrobot/sora2-watermark-remover

Removes the watermark from Sora 2 videos using a trained model and IOpaint

Public
6.6K runs

Run time and cost

This model costs approximately $0.049 to run on Replicate, or 20 runs per $1, but this varies depending on your inputs. It is also open source and you can run it on your own computer with Docker.

This model runs on Nvidia T4 GPU hardware. Predictions typically complete within 4 minutes. The predict time for this model varies significantly based on the inputs.

Readme

SoraWatermarkCleaner

This project provides an elegant way to remove the sora watermark in the sora2 generated videos.

  • Watermark removed

https://github.com/user-attachments/assets/8cdc075e-7d15-4d04-8fa2-53dd287e5f4c

  • Original

https://github.com/user-attachments/assets/3c850ff1-b8e3-41af-a46f-2c734406e77d

Method

The SoraWatermarkCleaner(we call it SoraWm later) is composed of two parts:

  • SoraWaterMarkDetector: We trained a yolov11s version to detect the sora watermark. (Thank you yolo!)

  • WaterMarkCleaner: We refer iopaint’s implementation for watermark removal using the lama model.

(This codebase is from https://github.com/Sanster/IOPaint#, thanks for their amazing work!)

Our SoraWm is purely deeplearning driven and yields good results in many generated videos.

License

Apache License

Citation

If you use this project, please cite:

@misc{sorawatermarkcleaner2025,
  author = {linkedlist771},
  title = {SoraWatermarkCleaner},
  year = {2025},
  url = {https://github.com/linkedlist771/SoraWatermarkCleaner}
}

Acknowledgments