1001 Demos

One Demo is Worth a Thousand Trajectories: Action-View Augmentation for Visuomotor Policies

We present 1001 Demos, an effective data augmentation framework that generates visually realistic fisheye image sequences and corresponding physically feasible action trajectories from real-world eye-in-hand demonstrations, captured with a portable parallel gripper with a single fisheye camera. We introduce a novel Gaussian Splatting formulation, adapted to wide FoV fisheye cameras, to reconstruct and edit the 3D scene with unseen objects. We utilize trajectory optimization to generate smooth, collision-free, view-renderingfriendly action trajectories and render visual observations from corresponding novel views. Comprehensive experiments in simulation and real world show that our augmentation framework improves the success rate for various manipulation tasks in both the same scene and the augmented scene with obstacles requiring collision avoidance.

Paper

Latest version: PDF.
Conference on Robot Learning (CoRL) 2025

Code and Resources

🔧 Code: Coming soon!

📊 Dataset: Coming soon!

Team

Chuer Pan¹

Litian Liang¹

Dominik Bauer²

Eric Cousineau³

Benjamin Burchfiel³

Siyuan Feng³

Shuran Song^1,2

¹ Stanford University           ² Columbia University           ³ Toyota Research Institute          

BibTeX

@inproceedings{pan20251001demos,
  title={One Demo is Worth a Thousand Trajectories: Action-View Augmentation for Visuomotor Policies},
  author={Pan, Chuer and Liang, Litian and Bauer, Dominik and Cousineau, Eric and Burchfiel, Benjamin and Feng, Siyuan and Song, Shuran},
  booktitle={Conference on Robot Learning (CoRL)},
  year={2025}
}