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

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Galappaththige, Chamuditha Jayanga, Lai, Jason, Windrim, Lloyd, Dansereau, Donald, Sünderhauf, Niko, Miller, Dimity
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908849414340608
author Galappaththige, Chamuditha Jayanga
Lai, Jason
Windrim, Lloyd
Dansereau, Donald
Sünderhauf, Niko
Miller, Dimity
author_facet Galappaththige, Chamuditha Jayanga
Lai, Jason
Windrim, Lloyd
Dansereau, Donald
Sünderhauf, Niko
Miller, Dimity
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2511_12370
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Changes in Real Time: Online Scene Change Detection with Multi-View Fusion
Galappaththige, Chamuditha Jayanga
Lai, Jason
Windrim, Lloyd
Dansereau, Donald
Sünderhauf, Niko
Miller, Dimity
Computer Vision and Pattern Recognition
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.
title Changes in Real Time: Online Scene Change Detection with Multi-View Fusion
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2511.12370