Improve model card: add metadata, paper, project page, code, and usage details

README.md

@@ -1 +1,114 @@
-arxiv.org/abs/2511.20614
+---
+license: cc-by-nc-4.0
+pipeline_tag: image-to-image
+library_name: diffusers
+---
+
+# The Consistency Critic: Correcting Inconsistencies in Generated Images via Reference-Guided Attentive Alignment
+
+This repository hosts **ImageCritic**, a reference-guided post-editing approach designed to correct inconsistencies in generated images. It aims to solve the inconsistency problem in generated images by applying attention alignment and a detail encoder, providing significant improvements over existing methods in various customized generation scenarios.
+
+The model was presented in the paper [The Consistency Critic: Correcting Inconsistencies in Generated Images via Reference-Guided Attentive Alignment](https://huggingface.co/papers/2511.20614).
+
+*   📚 [Paper (arXiv)](https://arxiv.org/abs/2511.20614)
+*   🌐 [Project Page](https://ouyangziheng.github.io/ImageCritic-Page/)
+*   💻 [Code (GitHub)](https://github.com/HVision-NKU/ImageCritic)
+*   🤗 [Hugging Face Space Demo](https://huggingface.co/spaces/ziheng1234/ImageCritic)
+*   📦 [Hugging Face Dataset](https://huggingface.co/datasets/ziheng1234/Critic-10K)
+
+<img src='https://github.com/HVision-NKU/ImageCritic/raw/main/figure/teaser.png' width='100%' />
+
+## 🖼️ Visual Results
+
+ImageCritic can effectively resolve detail-related issues in various customized generation scenarios, providing significant improvements over existing methods.
+
+<img src='https://github.com/HVision-NKU/ImageCritic/raw/main/figure/compare.png' width='100%' />
+
+## 🔧 Dependencies and Installation
+
+We recommend using Python 3.10 and PyTorch with CUDA support. To set up the environment:
+
+```bash
+# Create a new conda environment
+conda create -n imagecritic python=3.10
+conda activate imagecritic
+
+# Install other dependencies
+pip install -r requirements.txt
+```
+
+## ⚡ Quick Inference
+### Tips
+Due to copyright issues, we have embedded the download of the kontext model weights in the inference code below, You can run following inference code directly.
+If you have already downloaded the corresponding model, you can comment out the related code and directly replace the inference path.
+
+### Single case inference
+```bash
+python infer.py
+```
+
+### Local Gradio Demo
+```bash
+python app.py
+```
+
+### Single Model Download 
+You can download the base model FLUX.1-Kontext-dev directly from [Hugging Face](https://huggingface.co/black-forest-labs/FLUX.1-Kontext-dev).
+
+Alternatively, you can download it via the following command  
+(⚠️ Remember to replace `your_hf_token` in the script with your actual Hugging Face access token):
+```bash
+python ./download_kontext.py
+```
+
+You can download our ImageCritic directly from [Hugging Face](https://huggingface.co/ziheng1234/ImageCritic).
+
+Alternatively, you can download it via following code:
+
+```bash
+python ./download_imageCritic.py
+```
+Or using Git:
+```bash
+git lfs install
+git clone https://huggingface.co/ziheng1234/ImageCritic
+```
+
+## Dataset Download 
+
+You can download our training dataset Critic-10K directly from [Hugging Face](https://huggingface.co/datasets/ziheng1234/Critic-10K).
+
+Alternatively, you can download it via Python:
+
+```bash
+python /raid/users/oyzh/ImageCritic/download_dataset.py
+```
+Or using Git:
+```bash
+git lfs install
+git clone https://huggingface.co/datasets/ziheng1234/Critic-10K
+```
+
+### Online HuggingFace Demo
+You can try ImageCritic demo on [HuggingFace](https://huggingface.co/spaces/ziheng1234/ImageCritic).
+
+## Citation
+
+If ImageCritic is helpful, please help to ⭐ the repo.
+
+If you find this project useful for your research, please consider citing our paper:
+```bibtex
+@article{ouyang2025consistency,
+  title={The Consistency Critic: Correcting Inconsistencies in Generated Images via Reference-Guided Attentive Alignment},
+  author={Ouyang, Ziheng and Song, Yiren and Liu, Yaoli and Zhu, Shihao and Hou, Qibin and Cheng, Ming-Ming and Shou, Mike Zheng},
+  journal={arXiv preprint arXiv:2511.20614},
+  year={2025}
+}
+```
+
+## 📧 Contact
+If you have any comments or questions, please [open a new issue](https://github.com/HVision-NKU/ImageCritic/issues) or contact [Ziheng Ouyang](mailto:[email protected]) 
+
+## License
+Licensed under a [Creative Commons Attribution-NonCommercial 4.0 International](https://creativecommons.org/licenses/by-nc/4.0/) for Non-commercial use only.
+Any commercial use should get formal permission first.