File size: 2,677 Bytes
a6f2700 e9e8f89 a6f2700 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 | ---
library_name: openpi
license: other
license_name: gemma-terms
license_link: https://ai.google.dev/gemma/terms
pipeline_tag: robotics
---
<h1 align="center">ActiveScale: Scaling Active Perception for Robots<br>across Model, Data, and Hardware</h1>
<p align="center">
<a href="https://arxiv.org/abs/2609.18514"><img src="https://img.shields.io/badge/arXiv-B31B1B?style=for-the-badge&logo=arxiv&logoColor=white" alt="arXiv"></a>
<a href="https://active-scale.github.io/"><img src="https://img.shields.io/badge/Project_Page-4285F4?style=for-the-badge&logo=googlechrome&logoColor=white" alt="Project page"></a>
<a href="https://youtu.be/Iya0bSZ8Dko"><img src="https://img.shields.io/badge/YouTube-FF0000?style=for-the-badge&logo=youtube&logoColor=white" alt="YouTube"></a>
<a href="https://huggingface.co/davidzhou302/ActiveScale"><img src="https://img.shields.io/badge/Hugging_Face-FFD21E?style=for-the-badge&logo=huggingface&logoColor=black" alt="Hugging Face"></a>
<a href="https://modelscope.cn/datasets/shuai302/ActiveScale"><img src="https://img.shields.io/badge/ModelScope-624AFF?style=for-the-badge" alt="ModelScope"></a>
</p>
<p align="center">
<a href="https://youtu.be/Iya0bSZ8Dko">
<img src="https://img.youtube.com/vi/Iya0bSZ8Dko/maxresdefault.jpg" alt="ActiveScale video" width="100%">
</a>
</p>
<p align="center">
<a href="https://active-scale.github.io/">
<img src="https://raw.githubusercontent.com/active-scale/active-scale.github.io/main/assets/teaser_1.webp" alt="ActiveScale overview" width="100%">
</a>
</p>
<p align="center">
ActiveScale presents a scalable way to study active perception for robot manipulation across model, data, and hardware.<br>
It equips a vision-language-action model with temporal camera observations and camera-pose prediction,<br>
and introduces the Active-perception Mobile-manipulation Platform (AMP) for coordinated viewpoint and manipulation control.
</p>
<h2 align="center">Checkpoints</h2>
<table align="center">
<thead>
<tr><th>Checkpoint</th><th>Description</th></tr>
</thead>
<tbody>
<tr><td><code>midtraining</code></td><td>Mid-training checkpoint</td></tr>
<tr><td><code>bag</code></td><td>Bag task checkpoint</td></tr>
<tr><td><code>drawer</code></td><td>Drawer task checkpoint</td></tr>
<tr><td><code>pot</code></td><td>Pot task checkpoint</td></tr>
<tr><td><code>box</code></td><td>Box task checkpoint</td></tr>
<tr><td><code>under</code></td><td>Under-table task checkpoint</td></tr>
</tbody>
</table>
<p align="center">
Code and usage: <a href="https://github.com/ShuaiZhou302/ActiveScale">github.com/ShuaiZhou302/ActiveScale</a>
</p>
|