Grounding Predicates through Actions

TL;DR

This paper introduces a weakly supervised method for automatically labeling symbolic states in videos using action pre- and post-conditions, enabling efficient training of predicate classifiers for robotic reasoning.

Contribution

It presents a novel automatic labeling approach that reduces supervision costs and applies it to train predicate classifiers for symbolic reasoning in robotics.

Findings

Predicate classifiers match fully supervised performance

Automatic labeling significantly reduces annotation effort

Enables real-world task planning with learned predicates

Abstract

Symbols representing abstract states such as "dish in dishwasher" or "cup on table" allow robots to reason over long horizons by hiding details unnecessary for high-level planning. Current methods for learning to identify symbolic states in visual data require large amounts of labeled training data, but manually annotating such datasets is prohibitively expensive due to the combinatorial number of predicates in images. We propose a novel method for automatically labeling symbolic states in large-scale video activity datasets by exploiting known pre- and post-conditions of actions. This automatic labeling scheme only requires weak supervision in the form of an action label that describes which action is demonstrated in each video. We use our framework to train predicate classifiers to identify symbolic relationships between objects when prompted with object bounding boxes, and demonstrate that such predicate classifiers can match the performance of those trained with full supervision at a fraction of the labeling cost. We also apply our framework to an existing large-scale human activity dataset, and demonstrate the ability of these predicate classifiers trained on human data to enable closed-loop task planning in the real world.