lizhaoyang
commited on
Commit
·
5549bd5
1
Parent(s):
82e6abb
Add README
Browse files- .gitattributes +1 -0
- README.md +48 -0
- assets/figure1.png +3 -0
.gitattributes
CHANGED
|
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
| 36 |
+
*.png filter=lfs diff=lfs merge=lfs -text
|
README.md
CHANGED
|
@@ -1,3 +1,51 @@
|
|
| 1 |
---
|
| 2 |
license: apache-2.0
|
| 3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
license: apache-2.0
|
| 3 |
---
|
| 4 |
+
|
| 5 |
+
<h1 align="center">
|
| 6 |
+
BindWeave: Subject-Consistent Video Generation via Cross-Modal Integration
|
| 7 |
+
</h1>
|
| 8 |
+
|
| 9 |
+
|
| 10 |
+
<div align="center">
|
| 11 |
+
|
| 12 |
+
[](https://arxiv.org/pdf/2510.00438)
|
| 13 |
+
[](https://lzy-dot.github.io/BindWeave/)
|
| 14 |
+
<a href="https://huggingface.co/ByteDance/BindWeave"><img src="https://img.shields.io/static/v1?label=%F0%9F%A4%97%20Hugging%20Face&message=Model&color=orange"></a>
|
| 15 |
+
</div>
|
| 16 |
+
|
| 17 |
+
|
| 18 |
+
<p align="center">
|
| 19 |
+
<a href="https://arxiv.org/abs/2502.11079"><strong>BindWeave: Subject-Consistent Video Generation via Cross-Modal Integration</strong></a>
|
| 20 |
+
</p>
|
| 21 |
+
|
| 22 |
+
<div align="center">
|
| 23 |
+
<p>
|
| 24 |
+
<a href="https://scholar.google.com/citations?user=WelDcqkAAAAJ&hl=zh-CN">Zhaoyang Li</a><sup> 1,2</sup>,
|
| 25 |
+
<a href="https://openreview.net/profile?id=~Dongjun_Qian1">Dongjun Qian</a><sup> 2</sup>,
|
| 26 |
+
<a href="https://scholar.google.com/citations?user=Kp3XAToAAAAJ&hl=zh-CN">Kai Su</a><sup> 2*</sup>,
|
| 27 |
+
<a href="https://scholar.google.com/citations?user=G6xrfhYAAAAJ&hl=zh-CN">Qishuai Diao</a><sup> 2</sup>,
|
| 28 |
+
<a href="https://openreview.net/profile?id=~Xiangyang_Xia1">Xiangyang Xia</a><sup> 2</sup>,
|
| 29 |
+
<a href="https://openreview.net/profile?id=~Chang_Liu71">Chang Liu</a><sup> 2</sup>,
|
| 30 |
+
<a href="https://scholar.google.com/citations?user=rtO5VmQAAAAJ&hl=zh-CN">Wenfei Yang</a><sup> 1</sup>,
|
| 31 |
+
<a href="https://scholar.google.com/citations?user=9sCGe-gAAAAJ&hl=en">Tianzhu Zhang</a><sup> 1*</sup>,
|
| 32 |
+
<a href="https://shallowyuan.github.io/">Zehuan Yuan</a><sup> 2</sup>
|
| 33 |
+
</p>
|
| 34 |
+
<p>
|
| 35 |
+
<small>
|
| 36 |
+
<sup>1</sup>University of Science and Technology of China <sup>2</sup>ByteDance
|
| 37 |
+
<br>
|
| 38 |
+
<sup>*</sup>Corresponding Author
|
| 39 |
+
</small>
|
| 40 |
+
</p>
|
| 41 |
+
</div>
|
| 42 |
+
|
| 43 |
+
|
| 44 |
+
<p align="center">
|
| 45 |
+
<img src="assets/figure1.png" width=95%>
|
| 46 |
+
<p>
|
| 47 |
+
|
| 48 |
+
|
| 49 |
+
## 📖 Overview
|
| 50 |
+
BindWeave is a unified subject-consistent video generation framework for single- and multi-subject prompts, built on an MLLM-DiT architecture that couples a pretrained multimodal large language model with a diffusion transformer.
|
| 51 |
+
It achieves cross-modal integration via entity grounding and representation alignment, leveraging the MLLM to parse complex prompts and produce subject-aware hidden states that condition the DiT for high-fidelity generation.
|
assets/figure1.png
ADDED
|
Git LFS Details
|