arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2004.11085cs.CV
收藏

SL-DML:用于多模态一次性动作识别的信号级深度度量学习

SL-DML: Signal Level Deep Metric Learning for Multimodal One-Shot Action Recognition

University of Koblenz-Landau(科布伦茨-兰道大学)

Raphael Memmesheimer, Nick Theisen, Dietrich Paulus

摘要:

使用度量学习方法识别具有单个参考样本的活动是一个很有前途的研究领域。大多数Few-Shot方法集中在对象识别或面部识别上。我们提出了一种度量学习方法,将动作识别问题简化为嵌入空间中的最近邻搜索。我们将信号编码到图像中,并使用深度残差CNN提取特征。使用三重丢失,我们学习一个特征嵌入。由此产生的编码器将特征变换到嵌入空间中,在嵌入空间中,较近的距离编码相似的动作,而较高的距离编码不同的动作。我们的方法是基于一个信号水平的制定,并保持灵活的各种形式。它进一步优于大规模NTU RGB+D 120数据集上的One-Shot动作识别协议的基线5.6%。在只有60%的训练数据的情况下,我们的方法仍然比基线方法高出3.7%。对于40%的训练数据,我们的方法在第二次随访中表现良好。此外,我们表明,我们的方法在UTD-MHAD数据集的惯性,骨架和融合数据和模拟数据集的运动捕捉数据的实验中推广良好。此外,我们的关节间和传感器间实验表明,在以前看不见的设置良好的能力。

英文摘要:

Recognizing an activity with a single reference sample using metric learning approaches is a promising research field. The majority of few-shot methods focus on object recognition or face-identification. We propose a metric learning approach to reduce the action recognition problem to a nearest neighbor search in embedding space. We encode signals into images and extract features using a deep residual CNN. Using triplet loss, we learn a feature embedding. The resulting encoder transforms features into an embedding space in which closer distances encode similar actions while higher distances encode different actions. Our approach is based on a signal level formulation and remains flexible across a variety of modalities. It further outperforms the baseline on the large scale NTU RGB+D 120 dataset for the One-Shot action recognition protocol by 5.6%. With just 60% of the training data, our approach still outperforms the baseline approach by 3.7%. With 40% of the training data, our approach performs comparably well to the second follow up. Further, we show that our approach generalizes well in experiments on the UTD-MHAD dataset for inertial, skeleton and fused data and the Simitate dataset for motion capturing data. Furthermore, our inter-joint and inter-sensor experiments suggest good capabilities on previously unseen setups.

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。

↑