Learning to Answer Visual Questions from Web Videos

Result code in IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21730%2F25%3A00365339" target="_blank" >RIV/68407700:21730/25:00365339 - isvavai.cz</a>

Result on the web

<a href="https://doi.org/10.1109/TPAMI.2022.3173208" target="_blank" >https://doi.org/10.1109/TPAMI.2022.3173208</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1109/TPAMI.2022.3173208" target="_blank" >10.1109/TPAMI.2022.3173208</a>

Result language

angličtina

Original language name

Learning to Answer Visual Questions from Web Videos

Original language description

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. Furthermore, our method achieves competitive results on MSRVTT-QA, ActivityNet-QA, MSVD-QA and How2QA datasets. We also show that our approach generalizes to another source of web video and text data. We generate the WebVidVQA3M dataset from 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.

Czech name

—

Czech description

—

Type

J_imp - Article in a specialist periodical, which is included in the Web of Science database

CEP classification

—

OECD FORD branch

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Project

<a href="/en/project/EF15_003%2F0000468" target="_blank" >EF15_003/0000468: Intelligent Machine Perception</a><br>

Continuities

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Publication year

2025

Confidentiality

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Name of the periodical

IEEE Transactions on Pattern Analysis and Machine Intelligence

ISSN

0162-8828

e-ISSN

1939-3539

Volume of the periodical

47

Issue of the periodical within the volume

5

Country of publishing house

US - UNITED STATES

Number of pages

17

Pages from-to

3202-3218

UT code for WoS article

001465416300017

EID of the result in the Scopus database

2-s2.0-105002972696