# RePos: Relative-to-Absolute Pose Factorization for Cross-Environment WiFi-Based 3D Human Pose Estimation

Source: [RePos: Relative-to-Absolute Pose Factorization for Cross-Environment WiFi-Based 3D Human Pose Estimation](<https://www.sota2.com/research/paper/6a4c87dd85dba69fef8b233c>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-07-21T05:00:05.525Z

Authors: Zhangcheng Hou, Tomoaki Ohtsuki

Published: 2026-07-03T05:49:12.000Z

## Abstract

Device-free 3D human pose estimation from commodity WiFi Channel State Information (CSI) enables human sensing that preserves privacy and tolerates poor illumination, but its deployment is limited by poor generalization across environments. Unlike images, CSI measurements have no spatially localized correspondence to body parts and are heavily affected by multipath propagation. Consequently, models that regress absolute poses entangle body structure with location cues specific to each environment. Within a single environment this coupling is not problematic: RePos-D, a direct model that regresses the absolute pose, already achieves the best reported accuracy on Person-in-WiFi-3D, a 3.4% gain over the previous best WiFi method, DT-Pose. Across environments, however, the same model overfits position and degrades sharply. We therefore propose RePos, a factorized framework that separates root-relative pose estimation from root localization. By shielding the structure branch from absolute position, RePos learns robust pose representations. Specifically, it groups CSI features into latent tokens organized by body part that a skeleton-guided module refines into the pose, while a separate network estimates the root position from CSI amplitude through a differentiable spatial decomposition. Under the strict MM-Fi cross-environment protocol, RePos reduces the mean per-joint position error (MPJPE) by 10-21% over existing WiFi methods. The improvement is consistent across activity protocols, holds when each environment is held out in turn, and survives few-shot transfer without data leakage. Further analysis shows that the relative pose predictions remain largely independent of position, whereas root localization remains dependent on the environment.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [3D Human Pose Estimation on MM-Fi Protocol 1 14 daily activities E04 (test)](<https://www.sota2.com/research/sota/3d-human-pose-estimation-on-mm-fi-protocol-1-14-daily-activities-e04-test.md>): 3D Human Pose Estimation; MM-Fi Protocol 1 14 daily activities E04 (test); PCK@10: 4.9.

- [3D Human Pose Estimation on MM-Fi Protocol 2 13 rehabilitation activities E04 (test)](<https://www.sota2.com/research/sota/3d-human-pose-estimation-on-mm-fi-protocol-2-13-rehabilitation-activities-e04-test.md>): 3D Human Pose Estimation; MM-Fi Protocol 2 13 rehabilitation activities E04 (test); PCK@10: 2.9.

- [3D Human Pose Estimation on MM-Fi Protocol 3 27 activities E04 (test)](<https://www.sota2.com/research/sota/3d-human-pose-estimation-on-mm-fi-protocol-3-27-activities-e04-test.md>): 3D Human Pose Estimation; MM-Fi Protocol 3 27 activities E04 (test); PCK@10mm: 1.8.

- [3D Human Pose Estimation on Person-in-WiFi-3D single-environment (test)](<https://www.sota2.com/research/sota/3d-human-pose-estimation-on-person-in-wifi-3d-single-environment-test.md>): 3D Human Pose Estimation; Person-in-WiFi-3D single-environment (test); MPJPE: 86.9.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2607.02986>)

[PDF](<https://arxiv.org/pdf/2607.02986.pdf>)
