Saved in:
Bibliographic Details
Main Authors: Wu, Jingwei, Huang, Zhewei, Liu, Chang
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2505.19218
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915304181858304
author Wu, Jingwei
Huang, Zhewei
Liu, Chang
author_facet Wu, Jingwei
Huang, Zhewei
Liu, Chang
contents In the past decade, image foundation models (IFMs) have achieved unprecedented progress. However, the potential of directly using IFMs for video self-supervised representation learning has largely been overlooked. In this study, we propose an advancing video self-supervised learning (AdViSe) approach, aimed at significantly reducing the training overhead of video representation models using pre-trained IFMs. Specifically, we first introduce temporal modeling modules (ResNet3D) to IFMs, constructing a video representation model. We then employ a video self-supervised learning approach, playback rate perception, to train temporal modules while freezing the IFM components. Experiments on UCF101 demonstrate that AdViSe achieves performance comparable to state-of-the-art methods while reducing training time by $3.4\times$ and GPU memory usage by $8.2\times$. This study offers fresh insights into low-cost video self-supervised learning based on pre-trained IFMs. Code is available at https://github.com/JingwWu/advise-video-ssl.
format Preprint
id arxiv_https___arxiv_org_abs_2505_19218
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Advancing Video Self-Supervised Learning via Image Foundation Models
Wu, Jingwei
Huang, Zhewei
Liu, Chang
Computer Vision and Pattern Recognition
In the past decade, image foundation models (IFMs) have achieved unprecedented progress. However, the potential of directly using IFMs for video self-supervised representation learning has largely been overlooked. In this study, we propose an advancing video self-supervised learning (AdViSe) approach, aimed at significantly reducing the training overhead of video representation models using pre-trained IFMs. Specifically, we first introduce temporal modeling modules (ResNet3D) to IFMs, constructing a video representation model. We then employ a video self-supervised learning approach, playback rate perception, to train temporal modules while freezing the IFM components. Experiments on UCF101 demonstrate that AdViSe achieves performance comparable to state-of-the-art methods while reducing training time by $3.4\times$ and GPU memory usage by $8.2\times$. This study offers fresh insights into low-cost video self-supervised learning based on pre-trained IFMs. Code is available at https://github.com/JingwWu/advise-video-ssl.
title Advancing Video Self-Supervised Learning via Image Foundation Models
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2505.19218