RLIP: Relational Language-Image Pre-training for Human-Object Interaction Detection

TL;DR

RLIP introduces a novel contrastive pre-training strategy leveraging entity and relation descriptions, along with a new architecture and data augmentation techniques, to enhance human-object interaction detection in various learning scenarios.

Contribution

The paper proposes RLIP, a comprehensive pre-training framework with a new architecture, data generation, and noise mitigation methods for improved HOI detection.

Findings

Enhanced zero-shot and few-shot HOI detection performance.

Improved robustness to noisy annotations.

Significant gains in fine-tuning accuracy.

Abstract

The task of Human-Object Interaction (HOI) detection targets fine-grained visual parsing of humans interacting with their environment, enabling a broad range of applications. Prior work has demonstrated the benefits of effective architecture design and integration of relevant cues for more accurate HOI detection. However, the design of an appropriate pre-training strategy for this task remains underexplored by existing approaches. To address this gap, we propose Relational Language-Image Pre-training (RLIP), a strategy for contrastive pre-training that leverages both entity and relation descriptions. To make effective use of such pre-training, we make three technical contributions: (1) a new Parallel entity detection and Sequential relation inference (ParSe) architecture that enables the use of both entity and relation descriptions during holistically optimized pre-training; (2) a synthetic data generation framework, Label Sequence Extension, that expands the scale of language data available within each minibatch; (3) mechanisms to account for ambiguity, Relation Quality Labels and Relation Pseudo-Labels, to mitigate the influence of ambiguous/noisy samples in the pre-training data. Through extensive experiments, we demonstrate the benefits of these contributions, collectively termed RLIP-ParSe, for improved zero-shot, few-shot and fine-tuning HOI detection performance as well as increased robustness to learning from noisy annotations. Code will be available at https://github.com/JacobYuan7/RLIP.