Learning to Answer Visual Questions from Web Videos - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Journal Articles IEEE Transactions on Pattern Analysis and Machine Intelligence Year : 2022

Learning to Answer Visual Questions from Web Videos

Abstract

—Recent methods for visual question answering rely on large-scale annotated datasets. Manual annotation of questions and answers for videos, however, is tedious, expensive and prevents scalability. In this work, we propose to avoid manual annotation and generate a large-scale training dataset for video question answering making use of automatic cross-modal supervision. We leverage a question generation transformer trained on text data and use it to generate question-answer pairs from transcribed video narrations. Given narrated videos, we then automatically generate the HowToVQA69M dataset with 69M video-question-answer triplets. To handle the open vocabulary of diverse answers in this dataset, we propose a training procedure based on a contrastive loss between a video-question multi-modal transformer and an answer transformer. We introduce the zero-shot VideoQA task and the VideoQA feature probe evaluation setting and show excellent results, in particular for rare answers. Furthermore, our method achieves competitive results on MSRVTT-QA, ActivityNet-QA, MSVD-QA and How2QA datasets. We also show that our VideoQA dataset generation approach generalizes to another source of web video and text data. We use our method to generate the WebVidVQA3M dataset from the WebVid dataset, i.e., videos with alt-text annotations, and show its benefits for training VideoQA models. Finally, for a detailed evaluation we introduce iVQA, a new VideoQA dataset with reduced language bias and high-quality manual annotations. Code, datasets and trained models are available on our project webpage. Code, datasets and trained models are available at https://antoyang.github.io/just-ask.html.
Fichier principal
Vignette du fichier
TPAMIARXIV.pdf (6.44 Mo) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-03664182 , version 1 (10-05-2022)

Identifiers

Cite

Antoine Yang, Antoine Miech, Josef Sivic, Ivan Laptev, Cordelia Schmid. Learning to Answer Visual Questions from Web Videos. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2022, ⟨10.1109/tpami.2022.3173208⟩. ⟨hal-03664182⟩
61 View
92 Download

Altmetric

Share

Gmail Facebook X LinkedIn More