用图结构联合建模人体动作时空关系，提升3D姿态估计精度。
HSTGFormer: Hyper Spatial-Temporal Graph Transformer for 3D Human Pose Estimation
- 将时空推理转为关节-时间节点的局部图聚合，保持运动结构信息。
- 在Human3.6M和MPI-INF-3DHP上达到高精度，计算效率优异。
- 适合需要高效高精度3D姿态估计的视频分析场景。
基于Transformer的方法在单目3D人体姿态估计中表现强劲，但多数现有方法将空间与时间推理分阶段进行，削弱了人体运动中固有的统一时空依赖性，并在时间建模前压缩了帧级结构信息。本文提出HSTGFormer，一种图增强的Transformer框架，将时空推理重构为关节-时间节点上的局部耦合图聚合。具体而言，HSTGFormer引入超时空图（HSTG），通过将每帧骨架图扩展为时间邻域，将全局时空推理分解为个体关节-时间节点周围的局部时空感受野，从而实现结构感知的耦合推理并保留局部运动结构信息。同时，引入自适应双尺度时间图（ADSTG）以捕捉不同长度窗口下的关节特异性时间依赖。轻量级节点融合模块进一步自适应地整合两个图表示。在Human3.6M和MPI-INF-3DHP数据集上的实验表明，HSTGFormer在保持高精度的同时具备优秀的计算效率。
原文摘要 · Abstract (English)
Transformer-based methods have achieved strong performance in monocular 3D human pose estimation, but most existing approaches organise spatial and temporal reasoning as separate stages, which may weaken unified spatial-temporal interdependencies inherent in human motion and compress frame-level structural information before temporal modelling. In this paper, we propose HSTGFormer, a graph-enhanced Transformer framework that reformulates spatial-temporal reasoning as localised coupled graph aggregation over joint-time nodes. Specifically, HSTGFormer introduces a Hyper Spatial-Temporal Graph (HSTG), which decomposes global spatial-temporal reasoning into local spatial-temporal receptive fields around individual joint-time nodes by extending per-frame skeleton graphs into temporal neighbourhoods, thereby enabling structure-aware coupled reasoning while preserving local structural motion information. It further incorporates an Adaptive Dual-Scale Temporal Graph (ADSTG) to capture joint-specific temporal dependencies over complementary short- and long-range windows. A lightweight node-wise fusion module further adaptively integrates the two graph representations for each joint-time node. Experiments on Human3.6M and MPI-INF-3DHP show that HSTGFormer achieves strong accuracy with high computational efficiency.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。