mmFlux: Crowd Flow Analytics with Commodity mmWave MIMO Radar

2025-07-09

AI Summary

This work addresses the challenges of motion pattern modeling and semantic understanding in mmWave radar-based crowd sensing. We propose a novel method integrating optical flow principles with differential geometry analysis. First, radar echoes are reconstructed into high-fidelity 2D vector flow fields; then, directed geometric graphs are constructed, and local Jacobian matrices are employed to compute curl and divergence—enabling physically interpretable identification of collective semantic events (e.g., aggregation, dispersion, turning). To our knowledge, this is the first application of the optical flow estimation paradigm to mmWave signal processing, enhanced by morphological filtering and statistical denoising for robustness. Evaluated on 21 real-world experiments involving crowds of up to 20 people, the method achieves high-accuracy flow field reconstruction and semantic inference. Key metrics—including flow segmentation ratio, turning point localization, and boundary evolution—exhibit spatial alignment errors < 0.3 m, with quantitative performance significantly surpassing state-of-the-art approaches.

Technology Category

• Computer Vision: Motion & Tracking
• Planning, Routing, and Scheduling: Activity and Plan Recognition
• Humans and AI: Crowd Sourcing and Human Computation

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Security and Privacy: Large-scale security measurements
• Economics, Online Markets and Human Computation: Architectures and workflows that use LLMs for crowd work

Abstract

In this paper, we present a novel framework for extracting underlying crowd motion patterns and inferring crowd semantics using mmWave radar. First, our proposed signal processing pipeline combines optical flow estimation concepts from vision with novel statistical and morphological noise filtering to generate high-fidelity mmWave flow fields - compact 2D vector representations of crowd motion. We then introduce a novel approach that transforms these fields into directed geometric graphs, where edges capture dominant flow currents, vertices mark crowd splitting or merging, and flow distribution is quantified across edges. Finally, we show that by analyzing the local Jacobian and computing the corresponding curl and divergence, we can extract key crowd semantics for both structured and diffused crowds. We conduct 21 experiments on crowds of up to (and including) 20 people across 3 areas, using commodity mmWave radar. Our framework achieves high-fidelity graph reconstruction of the underlying flow structure, even for complex crowd patterns, demonstrating strong spatial alignment and precise quantitative characterization of flow split ratios. Finally, our curl and divergence analysis accurately infers key crowd semantics, e.g., abrupt turns, boundaries where flow directions shift, dispersions, and gatherings. Overall, these findings validate our framework, underscoring its potential for various crowd analytics applications.

Problem

• Extracting crowd motion patterns using mmWave radar
• Transforming flow fields into directed geometric graphs
• Inferring crowd semantics via curl and divergence analysis

Innovation

• Combines optical flow with noise filtering
• Transforms flow fields into directed graphs
• Analyzes Jacobian for crowd semantics