Linhui Xiao
commited on
Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,53 @@
|
|
| 1 |
-
---
|
| 2 |
-
license: apache-2.0
|
| 3 |
-
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
---
|
| 4 |
+
|
| 5 |
+
|
| 6 |
+
[//]: # (<br />)
|
| 7 |
+
<p align="center"> <h1 align="center">OneRef: Unified One-tower Expression Grounding and Segmentation with Mask Referring Modeling</h1>
|
| 8 |
+
<p align="center">
|
| 9 |
+
<b> NeurIPS 2024 </b>
|
| 10 |
+
<br />
|
| 11 |
+
<a href="https://scholar.google.com.hk/citations?user=4rTE4ogAAAAJ&hl=zh-CN&oi=sra"><strong> Linhui Xiao </strong></a>
|
| 12 |
+
路
|
| 13 |
+
<a href="https://yangxs.ac.cn/home"><strong>Xiaoshan Yang </strong></a>
|
| 14 |
+
路
|
| 15 |
+
<a href="https://scholar.google.com.hk/citations?user=HBZ9plsAAAAJ&hl=zh-CN"><strong>Fang Peng </strong></a>
|
| 16 |
+
路
|
| 17 |
+
<a href="https://scholar.google.com.hk/citations?user=o_DllmIAAAAJ&hl=zh-CN"><strong>Yaowei Wang </strong></a>
|
| 18 |
+
路
|
| 19 |
+
<a href="https://scholar.google.com.hk/citations?user=hI9NRDkAAAAJ&hl=zh-CN"><strong>Changsheng Xu</strong></a>
|
| 20 |
+
</p>
|
| 21 |
+
|
| 22 |
+
<p align="center">
|
| 23 |
+
<a href='https://openreview.net/pdf?id=siPdcro6uD'>
|
| 24 |
+
<img src='https://img.shields.io/badge/NeurIPS PDF-purple' alt='arXiv PDF'>
|
| 25 |
+
</a>
|
| 26 |
+
<a href='https://neurips.cc/virtual/2024/poster/93378'>
|
| 27 |
+
<img src='https://img.shields.io/badge/NeurIPS Paper Homepage-blue' alt='arXiv PDF'>
|
| 28 |
+
</a>
|
| 29 |
+
<a href='https://neurips.cc/media/PosterPDFs/NeurIPS%202024/93378.png?t=1729402553.3015864'>
|
| 30 |
+
<img src='https://img.shields.io/badge/NeurIPS Poster-lightblue' alt='arXiv PDF'>
|
| 31 |
+
</a>
|
| 32 |
+
<a href='https://neurips.cc/media/neurips-2024/Slides/93378_ROahXfO.pdf'>
|
| 33 |
+
<img src='https://img.shields.io/badge/NeurIPS Slides-lightgreen' alt='arXiv PDF'>
|
| 34 |
+
</a>
|
| 35 |
+
<a href='https://arxiv.org/pdf/2410.08021'>
|
| 36 |
+
<img src='https://img.shields.io/badge/arXiv-PDF-green?style=flat&logo=arXiv&logoColor=green' alt='arXiv PDF'>
|
| 37 |
+
</a>
|
| 38 |
+
<br />
|
| 39 |
+
|
| 40 |
+
|
| 41 |
+
|
| 42 |
+
<p align="center"> <img src='docs/fig1.jpg' align="center" width="95%"> </p>
|
| 43 |
+
|
| 44 |
+
**<p align="center"> A Comparison between OneRef model and the mainstream REC/RES architectures. </p>**
|
| 45 |
+
|
| 46 |
+
This repository is the official Pytorch implementation for the paper [**OneRef: Unified One-tower Expression Grounding
|
| 47 |
+
and Segmentation with Mask Referring Modeling**](https://openreview.net/pdf?id=siPdcro6uD), which is an advanced version
|
| 48 |
+
of our preliminary work **HiVG** ([Publication](https://dl.acm.org/doi/abs/10.1145/3664647.3681071), [Paper](https://openreview.net/pdf?id=NMMyGy1kKZ),
|
| 49 |
+
[Code](https://github.com/linhuixiao/HiVG)) and **CLIP-VG** ([Publication](https://ieeexplore.ieee.org/abstract/document/10269126),
|
| 50 |
+
[Paper](https://arxiv.org/pdf/2305.08685), [Code](https://github.com/linhuixiao/CLIP-VG)).
|
| 51 |
+
|
| 52 |
+
If you have any questions, please feel free to open an issue or contact me with emails: <xiaolinhui16@mails.ucas.ac.cn>.
|
| 53 |
+
Any kind discussions are welcomed!
|