• [03/27/2026] 🔥 arXiv paper is now available.
• [03/26/2026] 🔥 Demo is live, give it a try 🎮
• [03/25/2026] 🔥 Inference Code and Benchmark Data are released.
• [03/24/2026] 🔥 Project Page and Model Weight (Preview) are released.

• Project Page
• Model Weight (Preview)
• Inference Code
• Benchmark Data
• Online Demo
• Training Code
• Benchmark Code
• Model Weight (Stable)

Quick start for PixelSmile inference.

1. Install the environment in Installation.
2. Download the base model and PixelSmile weights in Model Download.
3. Run inference in Inference.

Clone the repository and enter the project directory:

git clone https://github.com/Ammmob/PixelSmile.git
cd PixelSmile

Create and activate a clean conda environment:

conda create -n pixelsmile python=3.10
conda activate pixelsmile

Install the inference dependencies:

pip install -r requirements.txt

Patch the current diffusers installation for the Qwen image edit bug:

bash scripts/patch_qwen_diffusers.sh

If you want to train PixelSmile, install the additional training dependencies on top of the inference environment:

pip install -r requirements-train.txt

PixelSmile uses Qwen/Qwen-Image-Edit-2511 as the base model.

✨ A more stable version is coming soon, with improved human expression editing performance and support for anime expression editing.

Training requires additional pretrained weights and auxiliary models. We will provide the full training asset list soon.

PixelSmile supports two simple ways to run inference.

bash scripts/run_infer.sh

You can edit scripts/run_infer.sh and directly modify the default values in DEFAULT_ARGS.

bash scripts/run_infer.sh \
  --image-path /path/to/input.jpg \
  --output-dir /path/to/output \
  --model-path /path/to/Qwen-Image-Edit-2511 \
  --lora-path /path/to/PixelSmile.safetensors \
  --expression happy \
  --scales 0 0.5 1.0 1.5 \
  --seed 42

Command-line arguments will override the default values in the script.

Training code is coming soon.

If you find PixelSmile useful in your research or applications, please consider citing our work.

@article{hua2026pixelsmile,
  title={PixelSmile: Toward Fine-Grained Facial Expression Editing},
  author={Jiabin Hua and Hengyuan Xu and Aojie Li and Wei Cheng and Gang Yu and Xingjun Ma and Yu-Gang Jiang},
  journal={arXiv preprint arXiv:2603.25728},
  year={2026}
}