3DGS-CD: 3D Gaussian Splatting-based Change Detection for Physical Object Rearrangement

TL;DR

This paper introduces 3DGS-CD, a novel method for detecting physical object rearrangements in 3D scenes using Gaussian Splatting, achieving high accuracy and speed without depth input or pre-defined object models.

Contribution

The paper presents the first 3D Gaussian Splatting-based approach for change detection, enabling fast, accurate, and model-free identification of object rearrangements in 3D environments.

Findings

Achieves up to 14% higher accuracy than state-of-the-art methods.

Operates with as few as one post-change image in 18 seconds.

Does not require depth input, user instructions, or pre-defined object classes.

Abstract

We present 3DGS-CD, the first 3D Gaussian Splatting (3DGS)-based method for detecting physical object rearrangements in 3D scenes. Our approach estimates 3D object-level changes by comparing two sets of unaligned images taken at different times. Leveraging 3DGS's novel view rendering and EfficientSAM's zero-shot segmentation capabilities, we detect 2D object-level changes, which are then associated and fused across views to estimate 3D change masks and object transformations. Our method can accurately identify changes in cluttered environments using sparse (as few as one) post-change images within as little as 18s. It does not rely on depth input, user instructions, pre-defined object classes, or object models -- An object is recognized simply if it has been re-arranged. Our approach is evaluated on both public and self-collected real-world datasets, achieving up to 14% higher accuracy and three orders of magnitude faster performance compared to the state-of-the-art radiance-field-based change detection method. This significant performance boost enables a broad range of downstream applications, where we highlight three key use cases: object reconstruction, robot workspace reset, and 3DGS model update. Our code and data will be made available at https://github.com/520xyxyzq/3DGS-CD.