Update README.md
Browse files
README.md
CHANGED
|
@@ -18,17 +18,23 @@ metrics:
|
|
| 18 |
- VBench
|
| 19 |
---
|
| 20 |
|
|
|
|
| 21 |
|
| 22 |
-
|
| 23 |
|
| 24 |
-
|
| 25 |
|
| 26 |
-
|
| 27 |
-
|
| 28 |
-
[[Paper (arXiv)]](https://arxiv.org/abs/2512.05044) · [[Project Page]](https://ivg-yanranzhang.github.io/MoGe4D/) · [[GitHub]](https://github.com/Zhangyr2022/MoGe4D) · [[Dataset (TrajScene-60K)]](https://www.modelscope.cn/datasets/YanranZhang/TrajScene-60K)
|
| 29 |
-
|
| 30 |
-
> Accepted to **ECCV 2026**.
|
| 31 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 32 |
## 📄 Paper Summary
|
| 33 |
|
| 34 |
Generating interactive and dynamic 4D scenes from a single static image is a core challenge. Existing methods decouple geometry from motion — either *generate-then-reconstruct* (geometric inconsistency) or *reconstruct-then-generate* (limited, externally-constrained motion) — causing spatiotemporal inconsistency and poor generalization.
|
|
|
|
| 18 |
- VBench
|
| 19 |
---
|
| 20 |
|
| 21 |
+
# [ECCV 2026] MoGe4D: Geometry-Aware Single-Image 4D Synthesis via Dense Trajectory Generation
|
| 22 |
|
| 23 |
+
<b>[Yanran Zhang](https://github.com/Zhangyr2022/)<sup>\*,1</sup>, [Ziyi Wang](https://wangzy22.github.io/)<sup>\*,1</sup>, [Wenzhao Zheng](https://wzzheng.net/#)<sup>†,1</sup>, [Zheng Zhu](http://www.zhengzhu.net/)<sup>2</sup>, [Jie Zhou](https://scholar.google.com/citations?user=6a79aPwAAAAJ&hl=en)<sup>1</sup>, [Jiwen Lu](https://ivg.au.tsinghua.edu.cn/Jiwen_Lu/)<sup>1</sup></b>
|
| 24 |
|
| 25 |
+
<sup>1</sup>Department of Automation, Tsinghua University <sup>2</sup>GigaAI
|
| 26 |
|
| 27 |
+
<i><sup>*</sup>Equal Contribution <sup>†</sup>Project Leader</i>
|
|
|
|
|
|
|
|
|
|
|
|
|
| 28 |
|
| 29 |
+
<p align="center">
|
| 30 |
+
<a href="https://github.com/Zhangyr2022/MoGe4D"><img src="https://img.shields.io/badge/GitHub-Code-black?logo=github" alt="GitHub"></a>
|
| 31 |
+
<a href="https://arxiv.org/abs/2512.05044"><img src="https://img.shields.io/badge/arXiv-Paper-b31b1b?logo=arxiv&logoColor=white" alt="arXiv"></a>
|
| 32 |
+
<a href="https://ivg-yanranzhang.github.io/MoGe4D/"><img src="https://img.shields.io/badge/Project-Website-blue?logo=googlechrome&logoColor=white" alt="Project"></a>
|
| 33 |
+
<br>
|
| 34 |
+
<a href="https://www.modelscope.cn/models/YanranZhang/MoGe4D"><img src="https://img.shields.io/badge/🤖%20ModelScope-Model-4e29ff" alt="ModelScope Model"></a>
|
| 35 |
+
<a href="https://huggingface.co/Yanran21/MoGe4D"><img src="https://img.shields.io/badge/🤗%20HuggingFace-Model-ffd21e" alt="HuggingFace Model"></a>
|
| 36 |
+
<a href="https://www.modelscope.cn/datasets/YanranZhang/TrajScene-60K"><img src="https://img.shields.io/badge/🤖%20ModelScope-Dataset-4e29ff" alt="ModelScope Dataset"></a>
|
| 37 |
+
</p>
|
| 38 |
## 📄 Paper Summary
|
| 39 |
|
| 40 |
Generating interactive and dynamic 4D scenes from a single static image is a core challenge. Existing methods decouple geometry from motion — either *generate-then-reconstruct* (geometric inconsistency) or *reconstruct-then-generate* (limited, externally-constrained motion) — causing spatiotemporal inconsistency and poor generalization.
|