Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Intensity Scan Context: Coding Intensity and Geometry Relations for Loop Closure Detection

About

Loop closure detection is an essential and challenging problem in simultaneous localization and mapping (SLAM). It is often tackled with light detection and ranging (LiDAR) sensor due to its view-point and illumination invariant properties. Existing works on 3D loop closure detection often leverage the matching of local or global geometrical-only descriptors, but without considering the intensity reading. In this paper we explore the intensity property from LiDAR scan and show that it can be effective for place recognition. Concretely, we propose a novel global descriptor, intensity scan context (ISC), that explores both geometry and intensity characteristics. To improve the efficiency for loop closure detection, an efficient two-stage hierarchical re-identification process is proposed, including a binary-operation based fast geometric relation retrieval and an intensity structure re-identification. Thorough experiments including both local experiment and public datasets test have been conducted to evaluate the performance of the proposed method. Our method achieves higher recall rate and recall precision than existing geometric-only methods.

Han Wang, Chen Wang, Lihua Xie• 2020

Related benchmarks

TaskDatasetResultRank
Place RecognitionKITTI Sequence 05
F1 Max77.1
9
Place RecognitionKITTI Sequence 08
F1 Score40.8
9
Place RecognitionKITTI Sequence 07
F1 max0.636
9
Place RecognitionKITTI Sequence 06
F1 max84.2
9
Place RecognitionKITTI Mean across sequences 00-08
F1 Max0.67
9
Place RecognitionKITTI Sequence 00
F1 max65.7
9
Place RecognitionKITTI Sequence 02
F1 Max70.5
9
Yaw EstimationKITTI (sequences 00, 02, 05, 06, 07, 08)
Yaw Error (Seq 00)0.829
4
Showing 8 of 8 rows

Other info

Follow for update