AeroDGS: Physically Consistent Dynamic Gaussian Splatting for Single-Sequence Aerial 4D Reconstruction

Researchers have developed AeroDGS, a physics-guided 4D Gaussian splatting framework that enables accurate dynamic scene reconstruction from single-view aerial UAV footage. The system addresses key challenges in monocular aerial reconstruction by incorporating physics-based optimization and geometric constraints to resolve depth ambiguity and improve motion estimation.

• →AeroDGS introduces a novel approach to reconstruct dynamic 3D scenes from single UAV camera sequences using physics-guided optimization.
• →The framework addresses critical limitations in aerial reconstruction including depth ambiguity, wide spatial ranges, and objects with large motion disparities.
• →A new Monocular Geometry Lifting module provides robust static and dynamic geometry reconstruction from aerial footage.
• →Physics-based constraints including ground-support, upright-stability, and trajectory-smoothness priors improve motion consistency.
• →Experimental results on both synthetic and real UAV datasets demonstrate superior performance compared to existing state-of-the-art methods.