Unconstrained Multi-view Human Pose Estimation with Algebraic Priors

TL;DR

This paper introduces a novel uncalibrated multi-view human pose estimation framework that combines neural networks, algebraic geometric priors, and temporal dynamics to achieve state-of-the-art results without camera calibration.

Contribution

It presents a data-driven triangulation method, an algebraic constraint loss, and a temporal coherence module, advancing uncalibrated multi-view human pose estimation.

Findings

Achieves new state-of-the-art performance on standard benchmarks.

Significantly narrows the gap between calibration-free and calibrated methods.

Demonstrates robustness in uncalibrated scenarios.

Abstract

Recovering 3D human pose from multi-view imagery typically relies on precise camera calibration, which is often unavailable in real-world scenarios, thereby severely limiting the applicability of existing methods. To overcome this challenge, we propose an unconstrained framework that synergizes deep neural networks, algebraic priors, and temporal dynamics for uncalibrated multi-view human pose estimation. First, we introduce the Triangulation with Transformer Regressor (TTR), which reformulates classical triangulation into a data-driven token fusion process to bypass the dependency on explicit camera parameters. Second, to explicitly embed the inherent algebraic relations of the multi-view variety into the learning process, we propose the Gr\"{o}bner basis Corrector (GC). This pioneering loss formulation enforces constraints derived from the multi-view variety to ensure the neural predictions strictly adhere to the laws of projective geometry. Finally, we devise the Temporal Equivariant Rectifier (TER), which exploits the equivariance property of human motion to impose temporal coherence and structural consistency, effectively mitigating scale ambiguity in uncalibrated settings. Extensive evaluations on standard benchmarks demonstrate that our framework establishes a new state-of-the-art for uncalibrated multi-view human pose estimation. Notably, our approach significantly closes the performance gap between calibration-free methods and fully calibrated oracles.