Long-Term Prediction of Local and Global Human Motion with Occlusion Recovery

2026-06-26

AI Summary

Existing human motion prediction methods predominantly rely on autoregressive frameworks, which are prone to error accumulation in long-term forecasting and struggle to jointly model local pose dynamics and global trajectory. Moreover, they exhibit limited robustness to occlusions or missing joints. To address these limitations, this work proposes a non-autoregressive Transformer architecture based on spatiotemporal attention that unifies the prediction of local poses and global motion within a single framework. The model accommodates variable-length historical inputs and enables reconstruction of missing joint observations. By circumventing the autoregressive constraint and leveraging multi-task training, the approach effectively mitigates error propagation, achieving significant performance gains in both long-horizon motion prediction and occlusion recovery tasks.

Abstract

Human motion describes the three-dimensional full-body movement of a person. Anticipating such motion holds significant relevance across a wide range of application domains such as human-robot interaction, autonomous driving, animation, and healthcare. In recent research, spatial and temporal dependencies are modeled by bidirectional attention mechanisms. These typically anticipate human motion in an autoregressive manner which could cause an accumulation of errors over time. As a consequence, they solely focus on local pose forecasting. To address these limitations, we propose a non-autoregressive transformer based on spatio-temporal attention, and train it not only for local pose anticipation, but also for global motion prediction in space. Furthermore, to enhance its applicability in real-world scenarios, our model is also trained to recover missing joints due to occlusions, and is capable of processing varying lengths of history observations. Our code is publicly available at https://github.com/Q-Y-Yang/Prediction-of-Local-and-Global-Human-Motion.

Problem

• human motion prediction
• long-term forecasting
• occlusion recovery
• global motion
• non-autoregressive modeling

Innovation

• non-autoregressive transformer
• spatio-temporal attention
• occlusion recovery
• global motion prediction
• long-term human motion forecasting