Hanyang Liu and Rongjun Qin introduced AeroDGS, a physics-guided 4D Gaussian splatting framework for monocular UAV videos, on February 25, 2026, addressing the challenges of depth ambiguity and unstable motion estimation in aerial scene reconstruction. The research presents a novel approach that reconstructs reliable static and dynamic geometry from single aerial sequences by incorporating differentiable physical constraints including ground-support, upright-stability, and trajectory-smoothness priors. This innovation transforms ambiguous visual data into physically consistent motion representations, solving a fundamental problem in computer vision where aerial conditions with wide spatial ranges and dynamic objects create inherently ill-posed reconstruction challenges. The framework jointly refines both static backgrounds and moving entities with stable geometry and coherent temporal evolution, significantly improving reconstruction fidelity in dynamic environments.
🏷️ Themes
Computer Vision, 3D Reconstruction, Aerial Imaging
A monocular is a compact refracting telescope used to magnify images of distant objects, typically using an optical prism to ensure an erect image, instead of using relay lenses like most telescopic sights. The volume and weight of a monocular are typically less than half of a pair of binoculars wi...
An unmanned aerial vehicle (UAV) or unmanned aircraft system (UAS), commonly known as a drone, is an aircraft with no human pilot, crew, or passengers on board, but rather is controlled remotely or is autonomous. UAVs were originally developed through the twentieth century for military missions too ...
Computer vision tasks include methods for acquiring, processing, analyzing, and understanding digital images, and extraction of high-dimensional data from the real world in order to produce numerical or symbolic information, e.g. in the form of decisions. "Understanding" in this context signifies th...
No entity connections available yet for this article.
Original Source
--> Computer Science > Computer Vision and Pattern Recognition arXiv:2602.22376 [Submitted on 25 Feb 2026] Title: AeroDGS: Physically Consistent Dynamic Gaussian Splatting for Single-Sequence Aerial 4D Reconstruction Authors: Hanyang Liu , Rongjun Qin View a PDF of the paper titled AeroDGS: Physically Consistent Dynamic Gaussian Splatting for Single-Sequence Aerial 4D Reconstruction, by Hanyang Liu and 1 other authors View PDF Abstract: Recent advances in 4D scene reconstruction have significantly improved dynamic modeling across various domains. However, existing approaches remain limited under aerial conditions with single-view capture, wide spatial range, and dynamic objects of limited spatial footprint and large motion disparity. These challenges cause severe depth ambiguity and unstable motion estimation, making monocular aerial reconstruction inherently ill-posed. To this end, we present AeroDGS, a physics-guided 4D Gaussian splatting framework for monocular UAV videos. AeroDGS introduces a Monocular Geometry Lifting module that reconstructs reliable static and dynamic geometry from a single aerial sequence, providing a robust basis for dynamic estimation. To further resolve monocular ambiguity, we propose a Physics-Guided Optimization module that incorporates differentiable ground-support, upright-stability, and trajectory-smoothness priors, transforming ambiguous image cues into physically consistent motion. The framework jointly refines static backgrounds and dynamic entities with stable geometry and coherent temporal evolution. We additionally build a real-world UAV dataset that spans various altitudes and motion conditions to evaluate dynamic aerial reconstruction. Experiments on synthetic and real UAV scenes demonstrate that AeroDGS outperforms state-of-the-art methods, achieving superior reconstruction fidelity in dynamic aerial environments. Comments: Accepted to CVPR 2026 Subjects: Computer Vision and Pattern Recognition (cs.CV) ; Artificial Intellige...
