arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2306.06010cs.CV
收藏

自监督视频表示学习的大规模分析

A Large-Scale Analysis on Self-Supervised Video Representation Learning

CRCV;University of Central Florida(中佛罗里达大学);BITS Pilani(pilani比特学院);Microsoft Research(微软研究院)

Akash Kumar, Ashlesha Kumar, Vibhav Vineet, Yogesh Singh Rawat

AI总结:

该研究构建了统一的视频自监督学习基准,系统分析了数据集规模、复杂度等五个关键方面,并提出仅需少量训练数据即可超越使用10倍预训练数据的SOTA方法。

摘要:

自监督学习是一种有效的无标签模型预训练方法,特别是在标注成本高昂的视频领域。现有视频领域的自监督工作使用不同的实验设置来展示其有效性,由于缺乏标准基准,跨方法的比较变得具有挑战性。在本研究中,我们首先提供了一个基准,使得能够在相同基础上对现有方法进行比较。接下来,我们研究了视频自监督学习中五个重要的不同方面:1)数据集规模,2)复杂度,3)数据分布,4)数据噪声,以及5)特征分析。为了促进这项研究,我们聚焦于七种不同的方法以及七种不同的网络架构,在5个不同的数据集上进行了广泛的实验,并评估了两个不同的下游任务。我们从这项研究中提出了几个有趣的见解,这些见解涵盖了预训练和目标数据集的不同属性、前置任务以及模型架构等。我们进一步将其中一些见解付诸真实测试,提出了一种仅需少量训练数据的方法,并且优于使用10倍预训练数据的现有最先进方法。我们相信这项工作将为研究人员更好地理解视频表示学习中的自监督前置任务铺平道路。

英文摘要:

Self-supervised learning is an effective way for label-free model pre-training, especially in the video domain where labeling is expensive. Existing self-supervised works in the video domain use varying experimental setups to demonstrate their effectiveness and comparison across approaches becomes challenging with no standard benchmark. In this work, we first provide a benchmark that enables a comparison of existing approaches on the same ground. Next, we study five different aspects of self-supervised learning important for videos; 1) dataset size, 2) complexity, 3) data distribution, 4) data noise, and, 5)feature analysis. To facilitate this study, we focus on seven different methods along with seven different network architectures and perform an extensive set of experiments on 5 different datasets with an evaluation of two different downstream tasks. We present several interesting insights from this study which span across different properties of pretraining and target datasets, pretext-tasks, and model architectures among others. We further put some of these insights to the real test and propose an approach that requires a limited amount of training data and outperforms existing state-of-the-art approaches which use 10x pretraining data. We believe this work will pave the way for researchers to a better understanding of self-supervised pretext tasks in video representation learning.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑