---
license: apache-2.0
pipeline_tag: image-to-video
tags:
- video-editing
- diffusion
- wan
---

<div align="center">

# NOVA: Sparse Control, Dense Synthesis for Pair-Free Video Editing

**CVPR 2026**

[![arXiv](https://img.shields.io/badge/arXiv-2603.02802-b31b1b)](https://arxiv.org/abs/2603.02802) [![GitHub](https://img.shields.io/badge/GitHub-NovaEdit-black?logo=github)](https://github.com/WeChatCV/NovaEdit)

</div>

![teaser](assets/teaser.jpg)

## Overview

NOVA is a pair-free video editing model built on **WAN 1.3B Fun InP**. It uses sparse keyframe control (e.g., a single edited first frame) to guide dense video synthesis, trained without requiring paired before/after video data.

- **Pair-free training** via degradation simulation
- **Sparse keyframe control**: provide one or more edited keyframes
- **Optional coarse mask** for improved editing accuracy

The framework consists of a sparse branch providing semantic guidance through user-edited keyframes and a dense branch that incorporates motion and texture information from the original video to maintain high fidelity and coherence.

## Usage

For full installation and training instructions, please visit the [GitHub repository](https://github.com/WeChatCV/NovaEdit).

### Inference via CLI

You can run inference using the `infer_nova.py` script. Below is an example for single GPU inference:

```bash
python infer_nova.py \
  --dataset_path ./example_videos \
  --metadata_file_name metadata.csv \
  --ckpt_path /path/to/checkpoints/stepXXX.ckpt \
  --output_path ./inference_results \
  --text_encoder_path /path/to/models_t5_umt5-xxl-enc-bf16.pth \
  --image_encoder_path /path/to/models_clip_open-clip-xlm-roberta-large-vit-huge-14.pth \
  --vae_path /path/to/Wan2.1_VAE.pth \
  --dit_path /path/to/diffusion_pytorch_model.safetensors \
  --num_samples 5 \
  --num_inference_steps 50 \
  --num_frames 81 \
  --height 480 \
  --width 832 \
  --first_only
```

## Citation

```bibtex
@article{pan2026nova,
  title={NOVA: Sparse Control, Dense Synthesis for Pair-Free Video Editing},
  author={Tianlin Pan and Jiayi Dai and Chenpu Yuan and Zhengyao Lv and Binxin Yang and Hubery Yin and Chen Li and Jing Lyu and Caifeng Shan and Chenyang Si},
  journal={arXiv preprint arXiv:2603.02802},
  year={2026}
}
```

## Acknowledgements

- [KlingTeam/ReCamMaster](https://github.com/KlingTeam/ReCamMaster)
- [zibojia/MiniMax-Remover](https://github.com/zibojia/MiniMax-Remover)