Audiovisual SlowFast Networks for Video Recognition

TL;DR

This paper introduces Audiovisual SlowFast Networks, a unified architecture for video recognition that integrates audio and visual data at multiple levels, achieving state-of-the-art results across various datasets.

Contribution

It proposes a novel integrated audiovisual network with hierarchical fusion and a DropPathway regularization technique to improve training and representation learning.

Findings

Achieves state-of-the-art results on six video datasets.

Demonstrates effective audiovisual feature learning in self-supervised settings.

Shows the generalization of the model to various video recognition tasks.

Abstract

We present Audiovisual SlowFast Networks, an architecture for integrated audiovisual perception. AVSlowFast has Slow and Fast visual pathways that are deeply integrated with a Faster Audio pathway to model vision and sound in a unified representation. We fuse audio and visual features at multiple layers, enabling audio to contribute to the formation of hierarchical audiovisual concepts. To overcome training difficulties that arise from different learning dynamics for audio and visual modalities, we introduce DropPathway, which randomly drops the Audio pathway during training as an effective regularization technique. Inspired by prior studies in neuroscience, we perform hierarchical audiovisual synchronization to learn joint audiovisual features. We report state-of-the-art results on six video action classification and detection datasets, perform detailed ablation studies, and show the generalization of AVSlowFast to learn self-supervised audiovisual features. Code will be made available at: https://github.com/facebookresearch/SlowFast.