Outside Knowledge Conversational Video (OKCV) Dataset -- Dialoguing over Videos

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Reichman, Benjamin, Patsch, Constantin, Truxal, Jack, Jain, Atishay, Heck, Larry
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866916790753296384
author Reichman, Benjamin
Patsch, Constantin
Truxal, Jack
Jain, Atishay
Heck, Larry
author_facet Reichman, Benjamin
Patsch, Constantin
Truxal, Jack
Jain, Atishay
Heck, Larry
contents In outside knowledge visual question answering (OK-VQA), the model must identify relevant visual information within an image and incorporate external knowledge to accurately respond to a question. Extending this task to a visually grounded dialogue setting based on videos, a conversational model must both recognize pertinent visual details over time and answer questions where the required information is not necessarily present in the visual information. Moreover, the context of the overall conversation must be considered for the subsequent dialogue. To explore this task, we introduce a dataset comprised of $2,017$ videos with $5,986$ human-annotated dialogues consisting of $40,954$ interleaved dialogue turns. While the dialogue context is visually grounded in specific video segments, the questions further require external knowledge that is not visually present. Thus, the model not only has to identify relevant video parts but also leverage external knowledge to converse within the dialogue. We further provide several baselines evaluated on our dataset and show future challenges associated with this task. The dataset is made publicly available here: https://github.com/c-patsch/OKCV.
format Preprint
id arxiv_https___arxiv_org_abs_2506_09953
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Outside Knowledge Conversational Video (OKCV) Dataset -- Dialoguing over Videos
Reichman, Benjamin
Patsch, Constantin
Truxal, Jack
Jain, Atishay
Heck, Larry
Computer Vision and Pattern Recognition
Artificial Intelligence
Computation and Language
In outside knowledge visual question answering (OK-VQA), the model must identify relevant visual information within an image and incorporate external knowledge to accurately respond to a question. Extending this task to a visually grounded dialogue setting based on videos, a conversational model must both recognize pertinent visual details over time and answer questions where the required information is not necessarily present in the visual information. Moreover, the context of the overall conversation must be considered for the subsequent dialogue. To explore this task, we introduce a dataset comprised of $2,017$ videos with $5,986$ human-annotated dialogues consisting of $40,954$ interleaved dialogue turns. While the dialogue context is visually grounded in specific video segments, the questions further require external knowledge that is not visually present. Thus, the model not only has to identify relevant video parts but also leverage external knowledge to converse within the dialogue. We further provide several baselines evaluated on our dataset and show future challenges associated with this task. The dataset is made publicly available here: https://github.com/c-patsch/OKCV.
title Outside Knowledge Conversational Video (OKCV) Dataset -- Dialoguing over Videos
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Computation and Language
url https://arxiv.org/abs/2506.09953