[CVPR 2025] ID-Patch: Robust ID Association for Group Photo Personalization

Yimeng Zhang^1,2,*, Tiancheng Zhi¹, Jing Liu¹, Shen Sang¹, Liming Jiang¹, Qing Yan¹, Sijia Liu², Linjie Luo^{1
  1}ByteDance Inc., ²Michigan State University
  ^*Work done during internship at ByteDance.

       

ID-Patch: Build Identity-to-Position Association

To address ID leakage and the linear increase in generation time with the number of identities, we propose ID-Patch, a novel method for robust identity-to-position association. From the same facial features, we generate both an ID patch—placed on the conditional image for precise spatial control—and ID embeddings, which are fused with text embeddings to enhance identity resemblance.

Environment Setup

Note: Python 3.9 and CUDA 12.2 are required. shell conda create -n idp python=3.9 conda activate idp pip install -r requirements.txt

Download models from https://huggingface.co/ByteDance/ID-Patch, and put them under models/ folder. shell git lfs install git clone https://huggingface.co/ByteDance/ID-Patch

Demo

shell python demo.py | Argument | Description | |----------|-------------| | --pose_image_path | Path to the pose image used for conditioning the generation. Default: data/poses/example_pose.png | | --subject_dir | Directory containing subject identity images. Each image should represent one person. Default: data/subjects | | --subjects | Comma-separated list of subject image filenames (e.g., exp_man.jpg,exp_woman.jpg). The order corresponds to their placement from left to right in the generated image. | | --prompt | Text prompt describing the scene to be generated. This guides the overall content and style of the output image. | | --base_model_path | Path to the base diffusion model to be used for generation. Default: RunDiffusion/Juggernaut-X-v10 | | --output_dir | Directory where the generated images will be saved. Default: results | | --output_name | Filename prefix for the generated image(s). Default: exp_result |

Disclaimer

Our released HuggingFace model differs from the paper’s version due to training on a different dataset.

License

``` Copyright 2024 Bytedance Ltd. and/or its affiliates

Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with the License. You may obtain a copy of the License at

http://www.apache.org/licenses/LICENSE-2.0

Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License. ```

Citation

If you find this code useful for your research, please cite us via the BibTeX below. BibTeX @InProceedings{zhang2025idpatch, author = {Zhang, Yimeng and Zhi, Tiancheng and Liu, Jing and Sang, Shen and Jiang, Liming and Yan, Qing and Liu, Sijia and Luo, Linjie}, title = {ID-Patch: Robust ID Association for Group Photo Personalization}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)}, month = {June}, year = {2025} }