TVLT: Textless Vision-Language Transformer

In this work, we present the Textless Vision-Language Transformer (TVLT), where homogeneous transformer blocks take raw visual and audio inputs for vision-and-language representation learning with minimal modality-specific design, and do not use text-specific modules such as tokenization or automati...

Full description

Saved in:
Bibliographic Details
Published in:arXiv.org 2022-11
Main Authors: Tang, Zineng, Cho, Jaemin, Nie, Yixin, Bansal, Mohit
Format: Article
Language:English
Subjects:
Audio signals
Automatic speech recognition
Data mining
Learning
Representations
Retrieval
Spectrograms
Transformers
Vision
Visual signals
Visual tasks
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In this work, we present the Textless Vision-Language Transformer (TVLT), where homogeneous transformer blocks take raw visual and audio inputs for vision-and-language representation learning with minimal modality-specific design, and do not use text-specific modules such as tokenization or automatic speech recognition (ASR). TVLT is trained by reconstructing masked patches of continuous video frames and audio spectrograms (masked autoencoding) and contrastive modeling to align video and audio. TVLT attains performance comparable to its text-based counterpart on various multimodal tasks, such as visual question answering, image retrieval, video retrieval, and multimodal sentiment analysis, with 28x faster inference speed and only 1/3 of the parameters. Our findings suggest the possibility of learning compact and efficient visual-linguistic representations from low-level visual and audio signals without assuming the prior existence of text. Our code and checkpoints are available at: https://github.com/zinengtang/TVLT
ISSN:2331-8422