arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2407.11048cs.LGcs.AI
收藏

幅度与旋转不变性下的交通方式检测:应对缺失数据模态

Magnitude and Rotation Invariant Detection of Transportation Modes with Missing Data Modalities

IDLab, Ghent University-imec(IDLab,根特大学-imec);Institute for Clarity in Documentation(文档清晰度研究所);Inria Paris-Rocquencourt(法国国家信息与自动化研究所巴黎-罗康分所);Rajiv Gandhi University(拉吉夫·甘地大学);Tsinghua University(清华大学);Palmer Research Laboratories(帕尔默研究实验室)

Jeroen Van Der Donckt, Jonas Van Der Donckt, Sofie Van Hoecke

AI总结:

针对2024年SHL挑战中数据缺失与分布偏移问题,提出幅度与旋转不变性聚合方法,显著优于传统特征,并缩短特征长度。

摘要:

本工作展示了Signal Sleuths团队在2024年SHL识别挑战赛中的解决方案。该挑战要求利用打乱顺序、不重叠的5秒窗口手机运动数据来检测交通方式,且三种可用模态(加速度计、陀螺仪、磁力计)中恰好有一种随机缺失。数据分析表明,训练数据与验证数据之间存在显著的分布偏移,因此需要采用幅度与旋转不变性的方法。我们采用传统机器学习,重点在于稳健的数据处理、特征提取以及旋转不变性聚合。消融研究表明,仅依赖常用的信号幅度向量会导致最差的性能。相反,我们提出的旋转不变性聚合相比使用旋转感知特征带来了显著改进,同时缩短了特征向量长度。此外,z归一化对于生成稳健的频谱特征至关重要。

英文摘要:

This work presents the solution of the Signal Sleuths team for the 2024 SHL recognition challenge. The challenge involves detecting transportation modes using shuffled, non-overlapping 5-second windows of phone movement data, with exactly one of the three available modalities (accelerometer, gyroscope, magnetometer) randomly missing. Data analysis indicated a significant distribution shift between train and validation data, necessitating a magnitude and rotation-invariant approach. We utilize traditional machine learning, focusing on robust processing, feature extraction, and rotation-invariant aggregation. An ablation study showed that relying solely on the frequently used signal magnitude vector results in the poorest performance. Conversely, our proposed rotation-invariant aggregation demonstrated substantial improvement over using rotation-aware features, while also reducing the feature vector length. Moreover, z-normalization proved crucial for creating robust spectral features.

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。

↑