RoboPCA: Pose-centered Affordance Learning from Human Demonstrations for Robot Manipulation

TL;DR

RoboPCA introduces a joint prediction framework for contact regions and poses in robot manipulation, utilizing a novel data collection pipeline from human demonstrations to improve task performance and generalization.

Contribution

The paper presents RoboPCA, a pose-centered affordance prediction method with a new data curation pipeline, Human2Afford, enabling scalable learning from human demonstrations.

Findings

Outperforms baseline methods on image datasets, simulation, and real robots.

Demonstrates strong generalization across tasks and object categories.

Effectively integrates geometry-appearance cues and mask features in a diffusion framework.

Abstract

Understanding spatial affordances -- comprising the contact regions of object interaction and the corresponding contact poses -- is essential for robots to effectively manipulate objects and accomplish diverse tasks. However, existing spatial affordance prediction methods mainly focus on locating the contact regions while delegating the pose to independent pose estimation approaches, which can lead to task failures due to inconsistencies between predicted contact regions and candidate poses. In this work, we propose RoboPCA, a pose-centered affordance prediction framework that jointly predicts task-appropriate contact regions and poses conditioned on instructions. To enable scalable data collection for pose-centered affordance learning, we devise Human2Afford, a data curation pipeline that automatically recovers scene-level 3D information and infers pose-centered affordance annotations from human demonstrations. With Human2Afford, scene depth and the interaction object's mask are extracted to provide 3D context and object localization, while pose-centered affordance annotations are obtained by tracking object points within the contact region and analyzing hand-object interaction patterns to establish a mapping from the 3D hand mesh to the robot end-effector orientation. By integrating geometry-appearance cues through an RGB-D encoder and incorporating mask-enhanced features to emphasize task-relevant object regions into the diffusion-based framework, RoboPCA outperforms baseline methods on image datasets, simulation, and real robots, and exhibits strong generalization across tasks and categories.