RAAP: Retrieval-Augmented Affordance Prediction with Cross-Image Action Alignment

TL;DR

RAAP is a novel framework that combines retrieval and alignment techniques to improve affordance prediction for robots, enabling better generalization and zero-shot manipulation.

Contribution

It introduces a unified retrieval-augmented alignment approach that decouples contact localization from action prediction, enhancing robustness and generalization.

Findings

RAAP achieves consistent performance on unseen objects and categories.

It enables zero-shot robotic manipulation in simulation and real-world settings.

Trained on small datasets, RAAP outperforms existing methods in affordance prediction.

Abstract

Understanding object affordances is essential for enabling robots to perform purposeful and fine-grained interactions in diverse and unstructured environments. However, existing approaches either rely on retrieval, which is fragile due to sparsity and coverage gaps, or on large-scale models, which frequently mislocalize contact points and mispredict post-contact actions when applied to unseen categories, thereby hindering robust generalization. We introduce Retrieval-Augmented Affordance Prediction (RAAP), a framework that unifies affordance retrieval with alignment-based learning. By decoupling static contact localization and dynamic action direction, RAAP transfers contact points via dense correspondence and predicts action directions through a retrieval-augmented alignment model that consolidates multiple references with dual-weighted attention. Trained on compact subsets of DROID and HOI4D with as few as tens of samples per task, RAAP achieves consistent performance across unseen objects and categories, and enables zero-shot robotic manipulation in both simulation and the real world. Project website: https://github.com/SEU-VIPGroup/RAAP.