Changes in Real Time: Online Scene Change Detection with Multi-View Fusion

TL;DR

This paper introduces a novel online scene change detection method that is pose-agnostic, label-free, and multi-view consistent, achieving real-time performance and surpassing state-of-the-art offline methods on complex datasets.

Contribution

The paper presents the first online SCD approach with multi-view fusion, self-supervised loss, and fast pose estimation, significantly improving accuracy and efficiency.

Findings

Operates at over 10 FPS with state-of-the-art accuracy.

Outperforms existing online and offline methods on real-world datasets.

Ensures multi-view consistency and pose-agnostic change detection.

Abstract

Online Scene Change Detection (SCD) is an extremely challenging problem that requires an agent to detect relevant changes on the fly while observing the scene from unconstrained viewpoints. Existing online SCD methods are significantly less accurate than offline approaches. We present the first online SCD approach that is pose-agnostic, label-free, and ensures multi-view consistency, while operating at over 10 FPS and achieving new state-of-the-art performance, surpassing even the best offline approaches. Our method introduces a new self-supervised fusion loss to infer scene changes from multiple cues and observations, PnP-based fast pose estimation against the reference scene, and a fast change-guided update strategy for the 3D Gaussian Splatting scene representation. Extensive experiments on complex real-world datasets demonstrate that our approach outperforms both online and offline baselines.