Home

Awesome

<div align="center"> <h1>A Simple Image Segmentation Framework via In-Context Examples </h1>

Yang Liu<sup>1</sup>,   Chenchen Jing<sup>1</sup>,   Hengtao Li<sup>1</sup>,   Muzhi Zhu<sup>1</sup>,   Hao Chen<sup>1</sup>,   Xinlong Wang<sup>2</sup>,   Chunhua Shen<sup>1</sup>

<sup>1</sup>Zhejiang University,   <sup>2</sup>Beijing Academy of Artificial Intelligence

NeurIPS 2024

</div>

🚀 Overview

<div align="center"> <img width="800" alt="image" src="figs/framework.png"> </div>

📖 Description

Overview

Paper

👻 Getting Started

🎫 License

For academic use, this project is licensed under the 2-clause BSD License. For commercial use, please contact Chunhua Shen.

🖊️ Citation

If you find this project useful in your research, please consider to cite:

@article{liu2024simple,
  title={A Simple Image Segmentation Framework via In-Context Examples},
  author={Liu, Yang and Jing, Chenchen and Li, Hengtao and Zhu, Muzhi and Chen, Hao and Wang, Xinlong and Shen, Chunhua},
  journal={Proc. Int. Conference on Neural Information Processing Systems (NeurIPS)},
  year={2024}
}

Acknowledgement

DINOv2, Mask2Former, SegGPT, Matcher, TFA and detectron2.

FAQ

Key Contributions of the Paper:

What is the main challenge in in-context segmentation that SINE aims to address?

How does SINE address task ambiguity?

How does SINE compare to SegGPT, another in-context segmentation model?

What are the limitations of SINE?