![[2602.18726] WiCompass: Oracle-driven Data Scaling for mmWave Human Pose Estimation](https://arxiv.org/static/browse/0.3.4/images/arxiv-logo-fb.png){kind=logo}
[2602.18726] WiCompass: Oracle-driven Data Scaling for mmWave Human Pose Estimation
Summary
The paper presents WiCompass, a framework for improving mmWave human pose estimation by focusing on data coverage rather than brute-force scaling, enhancing out-of-distribution robustness.
Why It Matters
This research addresses a significant challenge in mmWave human pose estimation, which is crucial for applications requiring privacy. By shifting the focus to coverage-aware data collection, it offers a more efficient method to enhance model performance in varied conditions, potentially impacting fields like robotics and surveillance.
Key Takeaways
- WiCompass improves robustness in mmWave human pose estimation.
- Focus on coverage-aware data collection enhances efficiency.
- The framework outperforms traditional brute-force data scaling methods.
- Utilizes large-scale motion-capture data to guide data collection.
- Addresses the issue of dataset redundancy effectively.
Computer Science > Computer Vision and Pattern Recognition arXiv:2602.18726 (cs) [Submitted on 21 Feb 2026] Title:WiCompass: Oracle-driven Data Scaling for mmWave Human Pose Estimation Authors:Bo Liang, Chen Gong, Haobo Wang, Qirui Liu, Rungui Zhou, Fengzhi Shao, Yubo Wang, Wei Gao, Kaichen Zhou, Guolong Cui, Chenren Xu View a PDF of the paper titled WiCompass: Oracle-driven Data Scaling for mmWave Human Pose Estimation, by Bo Liang and 10 other authors View PDF HTML (experimental) Abstract: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. Comments: Subjects: Comput...