├── LICENSE
├── README.md
└── assets
├── teaser.png
└── temporal_infer.png
/LICENSE:
--------------------------------------------------------------------------------
1 | MIT License
2 |
3 | Copyright (c) 2025 SpyderZSY
4 |
5 | Permission is hereby granted, free of charge, to any person obtaining a copy
6 | of this software and associated documentation files (the "Software"), to deal
7 | in the Software without restriction, including without limitation the rights
8 | to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
9 | copies of the Software, and to permit persons to whom the Software is
10 | furnished to do so, subject to the following conditions:
11 |
12 | The above copyright notice and this permission notice shall be included in all
13 | copies or substantial portions of the Software.
14 |
15 | THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
16 | IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
17 | FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
18 | AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
19 | LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
20 | OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
21 | SOFTWARE.
22 |
--------------------------------------------------------------------------------
/README.md:
--------------------------------------------------------------------------------
1 |
2 |
POMATO: Marrying Pointmap Matching with Temporal Motions
3 | for Dynamic 3D Reconstruction
4 |
5 |
6 |
7 |
8 |
9 | Songyan Zhang
1*, Yongtao Ge
2,3*, Jinyuan Tian
2*, Hao Chen
2†, Chen Lv
1, Chunhua Shen
2
10 |
11 |
1Nanyang Technology University, Singapore;
2Zhejiang University, China;
3The University of Adelaide, Australia
12 |
13 | *Equal Contributions, †Corresponding Author
14 |
15 |
16 |
17 | We present **POMATO** , a model that enables 3D reconstruction from an arbitrary dynamic video. Without relying on external modules, POMATO can
18 | directly perform 3D reconstruction along with temporal 3D point tracking and dynamic mask estimation.
19 |
20 | # Code will come soon!
21 |
22 | ## 🚀News
23 |
24 | - ```[Apr 2025]``` Released [paper](https://arxiv.org/abs/2504.05692) and init the github repo.
25 |
26 |
27 | ## 🔨 TODO LIST
28 |
29 | - [ ] Release the inference code and huggingface model.
30 | - [ ] Release the visualization of 3D tracking.
31 | - [ ] Release the training code.
32 |
33 | ## ✨Hightlights
34 | 🔥 We introduce a temporal motion module to facilitate the interactions of motion features along the temporal dimension.
35 |
36 |
37 |
38 |
39 |
40 | Inference pipelines for point tracking, video depth,
41 | and multi-view reconstruction with temporal motion module. tk indicates the keyframe in the
42 | sequence.
43 |
44 | ## 📌 Citation
45 |
46 | If you find our POMATO is useful in your research or applications, please consider giving a star ⭐ and citing using the following BibTeX:
47 |
48 | ```bibtex
49 | @article{zhang2025pomatomarryingpointmapmatching,
50 | title={POMATO: Marrying Pointmap Matching with Temporal Motion for Dynamic 3D Reconstruction},
51 | author={Songyan Zhang and Yongtao Ge and Jinyuan Tian and Guangkai Xu and Hao Chen and Chen Lv and Chunhua Shen},
52 | journal={arXiv preprint arXiv:2504.05692},
53 | year={2025},
54 | }
55 | ```
56 |
--------------------------------------------------------------------------------
/assets/teaser.png:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/wyddmw/POMATO/e875633b8706294011c12da1c7c47e329e50a4c0/assets/teaser.png
--------------------------------------------------------------------------------
/assets/temporal_infer.png:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/wyddmw/POMATO/e875633b8706294011c12da1c7c47e329e50a4c0/assets/temporal_infer.png
--------------------------------------------------------------------------------