Augmentron: Scalable Multi-View Visual Augmentation for Robot Learning
Augmentron: Scalable Multi-View Visual Augmentation for Robot Learning
This research introduces Augmentron, a geometry-aware pipeline that generates reusable scene representations and renders them through recorded camera poses to produce temporally and multi-view consistent environment variations while preserving robot, task objects, actions, and annotations. Replacing half of the real-world training dataset with Augmentron data improves model success across three physical manipulation tasks for both out-of-distribution distractor objects and workspace variation evaluations. In comparisons against alternate augmentation techniques, Augmentron achieves the highest pooled success in all three conditions, demonstrating that introducing augmented data to a training recipe helps policies generalize to dynamic real-world settings.
The paper proposes Augmentron, a geometry-aware pipeline for generating consistent multi-view visual augmentations to improve robot policy generalization on out-of-distribution distractors and workspace variations, with results reported across three physical manipulation tasks.
Source: arXiv Robotics — research abstracts · Read original article ↗
Loading article text…
What the source reports
Publisher-reported claims, with original evidence. These results have not been independently verified by RoboSignal.
- paper: not_reported
Source excerpts and review record
Automatically extracted; no manual editorial approval recorded.
introduce Augmentron, a geometry-aware pipeline that generates reusable scene representations and renders them through recorded camera poses. This produces temporally and multi-view consistent environ- ment and workspace variations while preserving the robot, task objects, actions, and annotations. Across three physical manipulation tasks, replacing half of the real world training dataset with Augmentron data improv
Open source S3
es model success across three tasks for both in out-of-distribution distractor objects, and out-of-distribution workspace variation evaluations. In a separate comparison against alternate augmentation techniques, Augmentron achieves the highest pooled success in all three conditions. The results show that introducing augmented data to a data training recipe can help policies generalize to dynamic real-world settings.
Open source S4
Augmentron: Scalable Multi-View Visual Augmentation for Robot Learning
Open source S1
arXiv:2611.09145v1 Announce Type: cross Abstract: Robot policies trained on visually narrow demonstrations can become sensitive to environmental features that are unrelated to the task. Existing generative augmentation methods can introduce visual diversity, but processing every frame and camera view independently is computationally expensive and may produce inconsistent observations across cameras and timestamps. We
Open source S2
Source：arXiv Robotics — research abstracts · arxiv.org