SpoxelNet: Spherical Voxel-based Deep Place Recognition for 3D Point Clouds of Crowded Indoor Spaces
| Autor: | Min Young Chang, Soo-Hyun Ryu, Suyong Yeon, Donghwan Lee |
|---|---|
| Rok vydání: | 2020 |
| Předmět: |
0209 industrial biotechnology
Computer science business.industry Deep learning Feature vector 020208 electrical & electronic engineering ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION Point cloud Spherical coordinate system 02 engineering and technology computer.software_genre 020901 industrial engineering & automation Voxel Margin (machine learning) 0202 electrical engineering electronic engineering information engineering Robot Computer vision Artificial intelligence business computer |
| Zdroj: | IROS |
| DOI: | 10.1109/iros45743.2020.9341549 |
| Popis: | With its essential role in achieving full autonomy of robot navigation, place recognition has been widely studied with various approaches. Recently, numerous point cloud-based methods with deep learning implementation have been proposed with promising results for their application in outdoor environments. However, their performances are not as promising in indoor spaces because of the high level of occlusion caused by structures and moving objects. In this paper, we propose a point cloud-based place recognition method for crowded indoor spaces. The method consists of voxelizing point clouds in spherical coordinates and defining the occupancy of each voxel in ternary values. We also present SpoxelNet, a neural network architecture that encodes input voxels into global descriptor vectors by extracting the structural features in both fine and coarse scales. It also reinforces its performance in occluded places by concatenating feature vectors from multiple directions. Our method is evaluated in various indoor datasets and outperforms existing methods with a large margin. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|