From Pixels to Primitives: Scene Change Detection in 3D Gaussian Splatting

TL;DR

This paper introduces GD-DIFF, a primitive-based scene change detection method using Gaussian splatting that improves multi-view consistency and change type identification without external models.

Contribution

It demonstrates that primitive attributes alone suffice for change detection and addresses Gaussian splatting ambiguities with geometric and photometric drift models.

Findings

GD-DIFF outperforms previous methods by ~17% in mean IoU on real-world benchmarks.

Operating on primitives yields inherently multi-view consistent change maps.

The method distinguishes structural from surface-level changes without supervision.

Abstract

Scene change detection methods built on Gaussian splatting universally follow a render-then-compare paradigm: the pre-change scene is rendered into 2D and compared against post-change images via pixel or feature residuals. This change detection problem with Gaussian Splatting has been treated as a question about pixels; we treat it as a question about primitives. We provide direct evidence that native primitive attributes alone -- position, anisotropic covariance, and color -- carry sufficient signal for scene change detection. What makes primitive-space comparison hard is the under-constrained nature of Gaussian splatting representation: independent optimizations yield primitive solutions whose count, positions, shapes, and colors differ even where nothing has changed. We address this challenge with anisotropic models of geometric and photometric drift, complemented by a per-primitive observability term that reflects the extent to which each Gaussian is constrained by the camera geometry. Operating directly on primitives gives our method, GD-DIFF, two properties that distinguish it from render-then-compare methods. First, change maps are multi-view consistent by construction, where prior work had to learn this through an additional optimization objective. Second, geometric and appearance changes are scored separately, identifying not just where but what kind of change occurred, distinguishing structural changes (e.g., an added object) from surface-level ones (e.g., a color change) without supervision or external model dependencies. On real-world benchmarks, GS-DIFF surpasses the prior state-of-the-art approach by $\sim$17% in mean Intersection over Union.