PRGCN: A Graph Memory Network for Cross-Sequence Pattern Reuse in 3D Human Pose Estimation

TL;DR

PRGCN introduces a graph memory network that reuses cross-sequence human pose patterns to improve 3D pose estimation accuracy and generalization, leveraging a pattern retrieval mechanism and hybrid spatiotemporal architecture.

Contribution

It proposes a novel pattern reuse framework with a graph memory bank and hybrid architecture, advancing 3D human pose estimation by exploiting cross-sequence regularities.

Findings

Achieves state-of-the-art MPJPE of 37.1mm on Human3.6M

Demonstrates improved cross-domain generalization

Effectively models pose patterns across sequences

Abstract

Monocular 3D human pose estimation remains a fundamentally ill-posed inverse problem due to the inherent depth ambiguity in 2D-to-3D lifting. While contemporary video-based methods leverage temporal context to enhance spatial reasoning, they operate under a critical paradigm limitation: processing each sequence in isolation, thereby failing to exploit the strong structural regularities and repetitive motion patterns that pervade human movement across sequences. This work introduces the Pattern Reuse Graph Convolutional Network (PRGCN), a novel framework that formalizes pose estimation as a problem of pattern retrieval and adaptation. At its core, PRGCN features a graph memory bank that learns and stores a compact set of pose prototypes, encoded as relational graphs, which are dynamically retrieved via an attention mechanism to provide structured priors. These priors are adaptively fused with hard-coded anatomical constraints through a memory-driven graph convolution, ensuring geometrical plausibility. To underpin this retrieval process with robust spatiotemporal features, we design a dual-stream hybrid architecture that synergistically combines the linear-complexity, local temporal modeling of Mamba-based state-space models with the global relational capacity of self-attention. Extensive evaluations on Human3.6M and MPI-INF-3DHP benchmarks demonstrate that PRGCN establishes a new state-of-the-art, achieving an MPJPE of 37.1mm and 13.4mm, respectively, while exhibiting enhanced cross-domain generalization capability. Our work posits that the long-overlooked mechanism of cross-sequence pattern reuse is pivotal to advancing the field, shifting the paradigm from per-sequence optimization towards cumulative knowledge learning.