Multi-View Pose-Agnostic Change Localization with Zero Labels
This addresses the need for accurate change detection in autonomous agents operating in dynamic, multi-view environments, with incremental improvements over existing methods.
The paper tackles the problem of detecting and localizing changes in environments from unconstrained viewpoints by proposing a label-free, pose-agnostic method that integrates multiple viewpoints into a change-aware 3D Gaussian Splatting representation, achieving state-of-the-art performance with a 1.7x improvement in Mean IoU and 1.5x in F1 score over baselines.
Autonomous agents often require accurate methods for detecting and localizing changes in their environment, particularly when observations are captured from unconstrained and inconsistent viewpoints. We propose a novel label-free, pose-agnostic change detection method that integrates information from multiple viewpoints to construct a change-aware 3D Gaussian Splatting (3DGS) representation of the scene. With as few as 5 images of the post-change scene, our approach can learn an additional change channel in a 3DGS and produce change masks that outperform single-view techniques. Our change-aware 3D scene representation additionally enables the generation of accurate change masks for unseen viewpoints. Experimental results demonstrate state-of-the-art performance in complex multi-object scenes, achieving a 1.7x and 1.5x improvement in Mean Intersection Over Union and F1 score respectively over other baselines. We also contribute a new real-world dataset to benchmark change detection in diverse challenging scenes in the presence of lighting variations.