Saved in:
Bibliographic Details
Main Authors: Qiao, Qianqian, Zheng, DanDan, Bo, Yihang, Peng, Bao, Huang, Heng, Jiang, Longteng, Wang, Huaye, Chen, Jingdong, Zhou, Jun, Jin, Xin
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2510.25238
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918200042586112
author Qiao, Qianqian
Zheng, DanDan
Bo, Yihang
Peng, Bao
Huang, Heng
Jiang, Longteng
Wang, Huaye
Chen, Jingdong
Zhou, Jun
Jin, Xin
author_facet Qiao, Qianqian
Zheng, DanDan
Bo, Yihang
Peng, Bao
Huang, Heng
Jiang, Longteng
Wang, Huaye
Chen, Jingdong
Zhou, Jun
Jin, Xin
contents Video aesthetic assessment, a vital area in multimedia computing, integrates computer vision with human cognition. Its progress is limited by the lack of standardized datasets and robust models, as the temporal dynamics of video and multimodal fusion challenges hinder direct application of image-based methods. This study introduces VADB, the largest video aesthetic database with 10,490 diverse videos annotated by 37 professionals across multiple aesthetic dimensions, including overall and attribute-specific aesthetic scores, rich language comments and objective tags. We propose VADB-Net, a dual-modal pre-training framework with a two-stage training strategy, which outperforms existing video quality assessment models in scoring tasks and supports downstream video aesthetic assessment tasks. The dataset and source code are available at https://github.com/BestiVictory/VADB.
format Preprint
id arxiv_https___arxiv_org_abs_2510_25238
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle VADB: A Large-Scale Video Aesthetic Database with Professional and Multi-Dimensional Annotations
Qiao, Qianqian
Zheng, DanDan
Bo, Yihang
Peng, Bao
Huang, Heng
Jiang, Longteng
Wang, Huaye
Chen, Jingdong
Zhou, Jun
Jin, Xin
Computer Vision and Pattern Recognition
Video aesthetic assessment, a vital area in multimedia computing, integrates computer vision with human cognition. Its progress is limited by the lack of standardized datasets and robust models, as the temporal dynamics of video and multimodal fusion challenges hinder direct application of image-based methods. This study introduces VADB, the largest video aesthetic database with 10,490 diverse videos annotated by 37 professionals across multiple aesthetic dimensions, including overall and attribute-specific aesthetic scores, rich language comments and objective tags. We propose VADB-Net, a dual-modal pre-training framework with a two-stage training strategy, which outperforms existing video quality assessment models in scoring tasks and supports downstream video aesthetic assessment tasks. The dataset and source code are available at https://github.com/BestiVictory/VADB.
title VADB: A Large-Scale Video Aesthetic Database with Professional and Multi-Dimensional Annotations
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2510.25238