Towards Arbitrary Motion Completing via Hierarchical Continuous Representation

2025-12-24

AI Summary

This paper addresses the problem of arbitrary frame-rate completion for human motion sequences and introduces Continuous Implicit Motion Representation (CIMR), the first framework enabling frame-rate-agnostic, high-fidelity interpolation, intermediate-frame synthesis, and long-term extrapolation. Methodologically, CIMR employs a hierarchical temporal encoding scheme to capture multi-scale motion dynamics and incorporates a Fourier-enhanced parametric activation function to improve the fitting capacity and generalization of implicit neural representations (INRs) for complex temporal patterns; the decoder adopts a lightweight MLP architecture. Evaluated on multiple standard benchmarks, CIMR achieves state-of-the-art performance in interpolation accuracy, extrapolation stability, temporal consistency, and cross-frame-rate generalization, while preserving motion smoothness and physical plausibility.

Technology Category

• Computer Vision: Motion & Tracking
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Physical motions are inherently continuous, and higher camera frame rates typically contribute to improved smoothness and temporal coherence. For the first time, we explore continuous representations of human motion sequences, featuring the ability to interpolate, inbetween, and even extrapolate any input motion sequences at arbitrary frame rates. To achieve this, we propose a novel parametric activation-induced hierarchical implicit representation framework, referred to as NAME, based on Implicit Neural Representations (INRs). Our method introduces a hierarchical temporal encoding mechanism that extracts features from motion sequences at multiple temporal scales, enabling effective capture of intricate temporal patterns. Additionally, we integrate a custom parametric activation function, powered by Fourier transformations, into the MLP-based decoder to enhance the expressiveness of the continuous representation. This parametric formulation significantly augments the model's ability to represent complex motion behaviors with high accuracy. Extensive evaluations across several benchmark datasets demonstrate the effectiveness and robustness of our proposed approach.

Problem

• Develops continuous representation for human motion sequences
• Enables interpolation and extrapolation at arbitrary frame rates
• Enhances motion smoothness and temporal coherence via hierarchical encoding

Innovation

• Hierarchical implicit representation for motion sequences
• Parametric activation with Fourier transformations
• Arbitrary frame rate interpolation and extrapolation