WiCompass: Oracle-driven Data Scaling for mmWave Human Pose Estimation
This work addresses robustness issues in privacy-preserving mmWave sensing for human pose estimation, offering a practical solution to data scaling bottlenecks.
The paper tackles the problem of poor generalization in millimeter-wave human pose estimation under distribution shifts by introducing WiCompass, a coverage-aware data-collection framework that uses an oracle to prioritize informative samples, resulting in improved out-of-distribution accuracy and superior scaling behavior compared to conventional strategies.
Millimeter-wave Human Pose Estimation (mmWave HPE) promises privacy but suffers from poor generalization under distribution shifts. We demonstrate that brute-force data scaling is ineffective for out-of-distribution (OOD) robustness; efficiency and coverage are the true bottlenecks. To address this, we introduce WiCompass, a coverage-aware data-collection framework. WiCompass leverages large-scale motion-capture corpora to build a universal pose space ``oracle'' that quantifies dataset redundancy and identifies underrepresented motions. Guided by this oracle, WiCompass employs a closed-loop policy to prioritize collecting informative missing samples. Experiments show that WiCompass consistently improves OOD accuracy at matched budgets and exhibits superior scaling behavior compared to conventional collection strategies. By shifting focus from brute-force scaling to coverage-aware data acquisition, this work offers a practical path toward robust mmWave sensing.