Audio-Adaptive Activity Recognition Across Video Domains

TL;DR

This paper introduces an audio-adaptive approach for activity recognition that leverages activity sounds to improve domain adaptation across different video settings, addressing challenges like actor and scenery changes.

Contribution

It proposes a novel audio-adaptive encoder and an audio-infused recognizer to enhance cross-domain activity recognition by utilizing less variable activity sounds.

Findings

Effective in reducing domain shift in activity recognition

Improves recognition accuracy on new datasets and scenarios

Addresses actor shift with a new dataset

Abstract

This paper strives for activity recognition under domain shift, for example caused by change of scenery or camera viewpoint. The leading approaches reduce the shift in activity appearance by adversarial training and self-supervised learning. Different from these vision-focused works we leverage activity sounds for domain adaptation as they have less variance across domains and can reliably indicate which activities are not happening. We propose an audio-adaptive encoder and associated learning methods that discriminatively adjust the visual feature representation as well as addressing shifts in the semantic distribution. To further eliminate domain-specific features and include domain-invariant activity sounds for recognition, an audio-infused recognizer is proposed, which effectively models the cross-modal interaction across domains. We also introduce the new task of actor shift, with a corresponding audio-visual dataset, to challenge our method with situations where the activity appearance changes dramatically. Experiments on this dataset, EPIC-Kitchens and CharadesEgo show the effectiveness of our approach.