PMGS: Reconstruction of Projectile Motion Across Large Spatiotemporal Spans via 3D Gaussian Splatting

TL;DR

This paper introduces PMGS, a novel method for reconstructing projectile motion over large spatiotemporal spans using 3D Gaussian Splatting, combining physics-based constraints and adaptive learning strategies.

Contribution

The paper presents a new framework for dynamic scene reconstruction that integrates physical constraints, dynamic annealing, and Kalman fusion for accurate projectile motion recovery.

Findings

Outperforms existing methods in high-speed nonlinear motion reconstruction

Effectively reduces error accumulation from multi-source observations

Achieves object-centric reconstruction with improved point density control

Abstract

Modeling complex rigid motion across large spatiotemporal spans remains an unresolved challenge in dynamic reconstruction. Existing paradigms are mainly confined to short-term, small-scale deformation and offer limited consideration for physical consistency. This study proposes PMGS, focusing on reconstructing Projectile Motion via 3D Gaussian Splatting. The workflow comprises two stages: 1) Target Modeling: achieving object-centralized reconstruction through dynamic scene decomposition and an improved point density control; 2) Motion Recovery: restoring full motion sequences by learning per-frame SE(3) poses. We introduce an acceleration consistency constraint to bridge Newtonian mechanics and pose estimation, and design a dynamic simulated annealing strategy that adaptively schedules learning rates based on motion states. Furthermore, we devise a Kalman fusion scheme to optimize error accumulation from multi-source observations to mitigate disturbances. Experiments show PMGS's superior performance in reconstructing high-speed nonlinear rigid motion compared to mainstream dynamic methods.