Update README.md
Browse files
README.md
CHANGED
@@ -1,3 +1,46 @@
|
|
1 |
-
---
|
2 |
-
license: apache-2.0
|
3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
pipeline_tag: mask-generation
|
4 |
+
base_model:
|
5 |
+
- OpenGVLab/InternVL2.5-4B
|
6 |
+
- facebook/sam2.1-hiera-large
|
7 |
+
tags:
|
8 |
+
- SeC
|
9 |
+
---
|
10 |
+
|
11 |
+
# SeC: Advancing Complex Video Object Segmentation via Progressive Concept Construction
|
12 |
+
|
13 |
+
[\[📂 GitHub\]](https://github.com/OpenIXCLab/SeC)
|
14 |
+
[\[📦 Benchmark\]](https://huggingface.co/datasets/OpenIXCLab/SeCVOS)
|
15 |
+
[\[🌐 Homepage\]](https://rookiexiong7.github.io/projects/SeC/)
|
16 |
+
[\[📄 Paper\]](https://arxiv.org/abs/2507.xxxxx)
|
17 |
+
|
18 |
+
## Highlights
|
19 |
+
|
20 |
+
- 🔥We introduce **Segment Concept (SeC)**, a **concept-driven** segmentation framework for **video object segmentation** that integrates **Large Vision-Language Models (LVLMs)** for robust, object-centric representations.
|
21 |
+
- 🔥SeC dynamically balances **semantic reasoning** with **feature matching**, adaptively adjusting computational efforts based on **scene complexity** for optimal segmentation performance.
|
22 |
+
- 🔥We propose the **Semantic Complex Scenarios Video Object Segmentation (SeCVOS)** benchmark, designed to evaluate segmentation in challenging scenarios.
|
23 |
+
|
24 |
+
## SeC Performance
|
25 |
+
|
26 |
+
| Model | SA-V val | SA-V test | LVOS v2 val | MOSE val | DAVIS 2017 val | YTVOS 2019 val | SeCVOS |
|
27 |
+
| :------ | :------: | :------: | :------: | :------: | :------: | :------: | :------: |
|
28 |
+
| SAM 2.1 | 78.6 | 79.6 | 84.1 | 74.5 | 90.6 | 88.7 | 58.2 |
|
29 |
+
| SAMURAI | 79.8 | 80.0 | 84.2 | 72.6 | 89.9 | 88.3 | 62.2 |
|
30 |
+
| SAM2.1Long | 81.1 | 81.2 | 85.9 | 75.2 | 91.4 | 88.7 | 62.3 |
|
31 |
+
| **SeC (Ours)** | **82.7** | **81.7** | **86.5** | **75.3** | **91.3** | **88.6** | **70.0** |
|
32 |
+
|
33 |
+
---
|
34 |
+
## Citation
|
35 |
+
|
36 |
+
If you find this project useful in your research, please consider citing:
|
37 |
+
|
38 |
+
```BibTeX
|
39 |
+
@article{zhang2025sec,
|
40 |
+
title = {SeC: Advancing Complex Video Object Segmentation via Progressive Concept Construction},
|
41 |
+
author = {Zhixiong Zhang and Shuangrui Ding and Xiaoyi Dong and Yuhang Zang and Yuhang Cao and Dahua Lin and Jiaqi Wang},
|
42 |
+
journal = {arXiv preprint arXiv:2507.xxxxx},
|
43 |
+
year = {2025},
|
44 |
+
url = {https://arxiv.org/abs/2507.xxxxx}
|
45 |
+
}
|
46 |
+
```
|