AVLnet: Learning Audio-Visual Language Representations from Instructional Videos

TL;DR

AVLnet is a self-supervised model that learns shared audio-visual representations directly from raw video inputs without relying on text annotations, achieving state-of-the-art results in retrieval tasks.

Contribution

The paper introduces AVLnet, a novel self-supervised approach for learning audio-visual embeddings from raw videos, and extends it with a tri-modal model incorporating text for enhanced retrieval.

Findings

Achieves state-of-the-art performance on image and video retrieval tasks.

Utilizes speech and natural sounds to learn meaningful audio-visual concepts.

Demonstrates effectiveness of self-supervised learning from raw video data.

Abstract

Current methods for learning visually grounded language from videos often rely on text annotation, such as human generated captions or machine generated automatic speech recognition (ASR) transcripts. In this work, we introduce the Audio-Video Language Network (AVLnet), a self-supervised network that learns a shared audio-visual embedding space directly from raw video inputs. To circumvent the need for text annotation, we learn audio-visual representations from randomly segmented video clips and their raw audio waveforms. We train AVLnet on HowTo100M, a large corpus of publicly available instructional videos, and evaluate on image retrieval and video retrieval tasks, achieving state-of-the-art performance. We perform analysis of AVLnet's learned representations, showing our model utilizes speech and natural sounds to learn audio-visual concepts. Further, we propose a tri-modal model that jointly processes raw audio, video, and text captions from videos to learn a multi-modal semantic embedding space useful for text-video retrieval. Our code, data, and trained models will be released at avlnet.csail.mit.edu