RigidFusion: Robot Localisation and Mapping in Environments with Large Dynamic Rigid Objects

TL;DR

This paper introduces RigidFusion, a novel RGB-D SLAM method that effectively segments, tracks, and reconstructs static and large dynamic rigid objects in environments with significant occlusion, without prior object knowledge.

Contribution

It treats all dynamic parts as one rigid body, enabling simultaneous localization and mapping of static and dynamic components without prior object information.

Findings

Improved segmentation of dynamic objects in challenging scenes

Enhanced camera localization accuracy amidst large occlusions

Successful reconstruction of both static and dynamic scene elements

Abstract

This work presents a novel RGB-D SLAM approach to simultaneously segment, track and reconstruct the static background and large dynamic rigid objects that can occlude major portions of the camera view. Previous approaches treat dynamic parts of a scene as outliers and are thus limited to a small amount of changes in the scene, or rely on prior information for all objects in the scene to enable robust camera tracking. Here, we propose to treat all dynamic parts as one rigid body and simultaneously segment and track both static and dynamic components. We, therefore, enable simultaneous localisation and reconstruction of both the static background and rigid dynamic components in environments where dynamic objects cause large occlusion. We evaluate our approach on multiple challenging scenes with large dynamic occlusion. The evaluation demonstrates that our approach achieves better motion segmentation, localisation and mapping without requiring prior knowledge of the dynamic object's shape and appearance.