RAM: Recover Any 3D Human Motion in-the-Wild

2026-03-20

AI Summary

This work addresses the challenge of maintaining identity consistency and motion coherence in 3D multi-person human motion reconstruction under severe occlusions and dynamic interactions in complex in-the-wild scenes. To this end, we propose the RAM framework, which integrates a motion-aware semantic tracker with an adaptive Kalman filter to achieve robust identity association. A memory-augmented temporal HMR module is introduced to inject spatiotemporal priors, while a lightweight pose predictor and a gated fusion mechanism effectively combine reconstructed and predicted features. Notably, our method is the first to jointly optimize adaptive Kalman filtering and temporal modeling, significantly enhancing tracking stability and 3D reconstruction accuracy in a zero-shot setting. Extensive experiments demonstrate that RAM outperforms state-of-the-art approaches on challenging in-the-wild benchmarks such as PoseTrack and 3DPW.

Technology Category

• Computer Vision: Motion & Tracking
• Intelligent Robots: State Estimation
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

RAM incorporates a motion-aware semantic tracker with adaptive Kalman filtering to achieve robust identity association under severe occlusions and dynamic interactions. A memory-augmented Temporal HMR module further enhances human motion reconstruction by injecting spatio-temporal priors for consistent and smooth motion estimation. Moreover, a lightweight Predictor module forecasts future poses to maintain reconstruction continuity, while a gated combiner adaptively fuses reconstructed and predicted features to ensure coherence and robustness. Experiments on in-the-wild multi-person benchmarks such as PoseTrack and 3DPW, demonstrate that RAM substantially outperforms previous state-of-the-art in both Zero-shot tracking stability and 3D accuracy, offering a generalizable paradigm for markerless 3D human motion capture in-the-wild.

Problem

• 3D human motion capture
• in-the-wild
• occlusion
• identity association
• motion reconstruction

Innovation

• motion-aware tracking
• adaptive Kalman filtering
• memory-augmented Temporal HMR
• pose prediction
• gated feature fusion