STCDiT: Spatio-Temporally Consistent Diffusion Transformer for High-Quality Video Super-Resolution

[Project Page]   [Paper]

Junyang Chen, Jiangxin Dong, Long Sun, Yixin Yang, Jinshan Pan,
IMAG Lab, Nanjing University of Science and Technology

If STCDiT is helpful for you, please help star the GitHub Repo. Thanks!

Welcome to visit our website (专注底层视觉领域的信息服务平台) for low-level vision: https://lowlevelcv.com/

---

🚩 New Features/Updates

• ✅ November 24, 2025. Create the repository.

⚡ To do

• Release the training code. Note that STCDiT-tiny can be trained on 4×24 GB GPUs with the same training settings as in paper.
• Release the Gradio Demo and ComfyUI Integration.
• Release the testing code and pre-trained model. Note that STCDiT-tiny can be inferred on a single 24 GB GPU.

📷 Real-World Enhancement Results

VideoLQ_024_video.mp4	VideoLQ_031_video.mp4
013_LQ.mp4	Sports_001_video.mp4
Sports_011_video.mp4	Sports_010_video.mp4

Contact

If you have any questions, please feel free to reach me out at [email protected].

---

Acknowledgments

Our project is based on DiffSynth-Studio and Wan 2.1. Thanks for their awesome works.