KHMP: Frequency-Domain Kalman Refinement for High-Fidelity Human Motion Prediction

TL;DR

KHMP introduces a frequency-domain Kalman filter with adaptive noise suppression and physical constraints to enhance the quality and realism of human motion predictions.

Contribution

The paper proposes KHMP, a novel framework combining adaptive Kalman filtering in the DCT domain with physics-based training constraints for improved motion prediction.

Findings

Achieves state-of-the-art accuracy on Human3.6M and HumanEva-I datasets.

Effectively reduces jitter and discontinuities in predicted motions.

Produces smooth, physically plausible human motion sequences.

Abstract

Stochastic human motion prediction aims to generate diverse, plausible futures from observed sequences. Despite advances in generative modeling, existing methods often produce predictions corrupted by high-frequency jitter and temporal discontinuities. To address these challenges, we introduce KHMP, a novel framework featuring an adaptiveKalman filter applied in the DCT domain to generate high-fidelity human motion predictions. By treating high-frequency DCT coefficients as a frequency-indexed noisy signal, the Kalman filter recursively suppresses noise while preserving motion details. Notably, its noise parameters are dynamically adjusted based on estimated Signal-to-Noise Ratio (SNR), enabling aggressive denoising for jittery predictions and conservative filtering for clean motions. This refinement is complemented by training-time physical constraints (temporal smoothness and joint angle limits) that encode biomechanical principles into the generative model. Together, these innovations establish a new paradigm integrating adaptive signal processing with physics-informed learning. Experiments on the Human3.6M and HumanEva-I datasets demonstrate that KHMP achieves state-of-the-art accuracy, effectively mitigating jitter artifacts to produce smooth and physically plausible motions.