arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2404.06243cs.CVcs.AIcs.HCcs.LGcs.MM
收藏

ActNetFormer:用于视频半监督动作识别的Transformer-ResNet混合方法

ActNetFormer: Transformer-ResNet Hybrid Method for Semi-Supervised Action Recognition in Videos

School of Information Technology, Monash University(莫纳什大学信息技术学院);TCS Research(塔塔咨询服务研究中心)

Sharana Dharshikgan Suresh Dass, Hrishav Bakul Barua, Ganesh Krishnasamy, Raveendran Paramesran, Raphael C. -W. Phan

AI总结:

本文提出ActNetFormer框架,通过跨架构伪标签与对比学习融合3D CNN和视频Transformer,仅用少量标注数据即在半监督视频动作识别中取得最先进性能。

摘要:

视频中的人类动作或活动识别是计算机视觉中的一项基础任务,在监控与监测、自动驾驶汽车、体育分析、人机交互等众多领域均有应用。传统的监督方法需要大量标注数据集进行训练,而获取这些数据既昂贵又耗时。本工作提出了一种新颖的方法,即采用跨架构伪标签与对比学习相结合的半监督动作识别方法。我们的框架同时利用有标签和无标签数据来稳健地学习视频中的动作表示,将伪标签与对比学习相结合,从而从两类样本中有效学习。我们引入了一种新颖的跨架构方法,利用3D卷积神经网络(3D CNN)和视频Transformer(VIT)来捕捉动作表示的不同方面;因此我们将其命名为ActNetFormer。3D CNN擅长捕捉空间特征和时间域中的局部依赖关系,而VIT擅长捕捉跨帧的远程依赖关系。通过在ActNetFormer框架内整合这些互补架构,我们的方法能够有效捕捉动作的局部和全局上下文信息。这种全面的表示学习使模型能够利用每种架构的优势,在半监督动作识别任务中取得更好的性能。在标准动作识别数据集上的实验结果表明,我们的方法优于现有方法,仅使用少量标注数据就达到了最先进的性能。本工作的官方网站可在以下地址访问:https://github.com/rana2149/ActNetFormer。

英文摘要:

Human action or activity recognition in videos is a fundamental task in computer vision with applications in surveillance and monitoring, self-driving cars, sports analytics, human-robot interaction and many more. Traditional supervised methods require large annotated datasets for training, which are expensive and time-consuming to acquire. This work proposes a novel approach using Cross-Architecture Pseudo-Labeling with contrastive learning for semi-supervised action recognition. Our framework leverages both labeled and unlabelled data to robustly learn action representations in videos, combining pseudo-labeling with contrastive learning for effective learning from both types of samples. We introduce a novel cross-architecture approach where 3D Convolutional Neural Networks (3D CNNs) and video transformers (VIT) are utilised to capture different aspects of action representations; hence we call it ActNetFormer. The 3D CNNs excel at capturing spatial features and local dependencies in the temporal domain, while VIT excels at capturing long-range dependencies across frames. By integrating these complementary architectures within the ActNetFormer framework, our approach can effectively capture both local and global contextual information of an action. This comprehensive representation learning enables the model to achieve better performance in semi-supervised action recognition tasks by leveraging the strengths of each of these architectures. Experimental results on standard action recognition datasets demonstrate that our approach performs better than the existing methods, achieving state-of-the-art performance with only a fraction of labeled data. The official website of this work is available at: https://github.com/rana2149/ActNetFormer.

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。

↑