City-scale Scene Change Detection using Point Clouds

TL;DR

This paper introduces a deep learning-based non-rigid registration method for detecting structural changes in city-scale scenes using point clouds generated from vehicular images, overcoming geo-location inaccuracies.

Contribution

It presents a novel deep learning approach for non-rigid registration of point clouds to improve change detection accuracy in urban environments.

Findings

Effective detection of scene changes despite viewpoint differences

Robustness to illumination variations demonstrated in experiments

Dual thresholding enhances change detection reliability

Abstract

We propose a method for detecting structural changes in a city using images captured from vehicular mounted cameras over traversals at two different times. We first generate 3D point clouds for each traversal from the images and approximate GNSS/INS readings using Structure-from-Motion (SfM). A direct comparison of the two point clouds for change detection is not ideal due to inaccurate geo-location information and possible drifts in the SfM. To circumvent this problem, we propose a deep learning-based non-rigid registration on the point clouds which allows us to compare the point clouds for structural change detection in the scene. Furthermore, we introduce a dual thresholding check and post-processing step to enhance the robustness of our method. We collect two datasets for the evaluation of our approach. Experiments show that our method is able to detect scene changes effectively, even in the presence of viewpoint and illumination differences.