Enhanced Sparse Point Cloud Data Processing for Privacy-aware Human Action Recognition

2025-08-14

AI Summary

Millimeter-wave radar sparse point clouds for privacy-preserving human activity recognition (HAR) suffer from low point density, high noise, and discontinuous trajectories. Method: This paper presents the first systematic evaluation of three core preprocessing techniques—DBSCAN clustering, Hungarian algorithm-based inter-frame association, and Kalman filter-based trajectory prediction—assessing both individual and full combinatorial performance. Experiments are conducted end-to-end on the MiliPoint dataset. We propose enhanced strategies: adaptive DBSCAN neighborhood parameter selection, motion-constrained Hungarian matching, and robust Kalman observation updates. Contribution/Results: The optimal pipeline improves HAR accuracy by 12.7%, significantly enhancing noise suppression and motion continuity modeling. Moreover, we quantitatively characterize the accuracy–computational cost trade-offs for each method—marking the first such analysis—thereby providing empirical evidence and methodological guidance for practical radar-based HAR deployment.

Technology Category

• Planning, Routing, and Scheduling: Activity and Plan Recognition
• Computer Vision: Motion & Tracking
• Machine Learning: Hardware-aware ML

Application Category

• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Security and Privacy: Large-scale security measurements
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics

Abstract

Human Action Recognition (HAR) plays a crucial role in healthcare, fitness tracking, and ambient assisted living technologies. While traditional vision based HAR systems are effective, they pose privacy concerns. mmWave radar sensors offer a privacy preserving alternative but present challenges due to the sparse and noisy nature of their point cloud data. In the literature, three primary data processing methods: Density-Based Spatial Clustering of Applications with Noise (DBSCAN), the Hungarian Algorithm, and Kalman Filtering have been widely used to improve the quality and continuity of radar data. However, a comprehensive evaluation of these methods, both individually and in combination, remains lacking. This paper addresses that gap by conducting a detailed performance analysis of the three methods using the MiliPoint dataset. We evaluate each method individually, all possible pairwise combinations, and the combination of all three, assessing both recognition accuracy and computational cost. Furthermore, we propose targeted enhancements to the individual methods aimed at improving accuracy. Our results provide crucial insights into the strengths and trade-offs of each method and their integrations, guiding future work on mmWave based HAR systems

Problem

• Evaluating radar data methods for privacy-aware action recognition
• Enhancing accuracy of sparse mmWave point cloud processing
• Analyzing trade-offs between recognition accuracy and computational cost

Innovation

• Combines DBSCAN, Hungarian Algorithm, Kalman Filtering
• Enhances mmWave radar sparse point cloud processing
• Evaluates performance and computational cost trade-offs