Pose-aware Multi-level Feature Network for Human Object Interaction Detection

TL;DR

This paper introduces a pose-aware multi-level feature network that leverages human pose cues and semantic part information to improve the detection of human-object interactions in complex scenes, achieving superior accuracy.

Contribution

It presents a novel multi-branch deep network utilizing pose cues at three semantic levels for more accurate and interpretable human-object interaction detection.

Findings

Outperforms prior methods on public benchmarks

Handles complex scenes with diverse human-object configurations

Provides interpretable interaction predictions

Abstract

Reasoning human object interactions is a core problem in human-centric scene understanding and detecting such relations poses a unique challenge to vision systems due to large variations in human-object configurations, multiple co-occurring relation instances and subtle visual difference between relation categories. To address those challenges, we propose a multi-level relation detection strategy that utilizes human pose cues to capture global spatial configurations of relations and as an attention mechanism to dynamically zoom into relevant regions at human part level. Specifically, we develop a multi-branch deep network to learn a pose-augmented relation representation at three semantic levels, incorporating interaction context, object features and detailed semantic part cues. As a result, our approach is capable of generating robust predictions on fine-grained human object interactions with interpretable outputs. Extensive experimental evaluations on public benchmarks show that our model outperforms prior methods by a considerable margin, demonstrating its efficacy in handling complex scenes.