A Multi-Domain Feature Learning Method for Visual Place Recognition
Visual Place Recognition (VPR) is an important component in both computer vision and robotics applications, thanks to its ability to determine whether a place has been visited and where specifically. A major challenge in VPR is to handle changes of environmental conditions including weather, season...
Main Authors: | , , , , , , , , |
---|---|
Format: | Report |
Language: | unknown |
Published: |
arXiv
2019
|
Subjects: | |
Online Access: | https://dx.doi.org/10.48550/arxiv.1902.10058 https://arxiv.org/abs/1902.10058 |
id |
ftdatacite:10.48550/arxiv.1902.10058 |
---|---|
record_format |
openpolar |
spelling |
ftdatacite:10.48550/arxiv.1902.10058 2023-05-15T17:24:38+02:00 A Multi-Domain Feature Learning Method for Visual Place Recognition Yin, Peng Xu, Lingyun Li, Xueqian Yin, Chen Li, Yingli Srivatsan, Rangaprasad Arun Li, Lu Ji, Jianmin He, Yuqing 2019 https://dx.doi.org/10.48550/arxiv.1902.10058 https://arxiv.org/abs/1902.10058 unknown arXiv arXiv.org perpetual, non-exclusive license http://arxiv.org/licenses/nonexclusive-distrib/1.0/ Robotics cs.RO Artificial Intelligence cs.AI Computer Vision and Pattern Recognition cs.CV FOS Computer and information sciences Preprint Article article CreativeWork 2019 ftdatacite https://doi.org/10.48550/arxiv.1902.10058 2022-04-01T08:47:12Z Visual Place Recognition (VPR) is an important component in both computer vision and robotics applications, thanks to its ability to determine whether a place has been visited and where specifically. A major challenge in VPR is to handle changes of environmental conditions including weather, season and illumination. Most VPR methods try to improve the place recognition performance by ignoring the environmental factors, leading to decreased accuracy decreases when environmental conditions change significantly, such as day versus night. To this end, we propose an end-to-end conditional visual place recognition method. Specifically, we introduce the multi-domain feature learning method (MDFL) to capture multiple attribute-descriptions for a given place, and then use a feature detaching module to separate the environmental condition-related features from those that are not. The only label required within this feature learning pipeline is the environmental condition. Evaluation of the proposed method is conducted on the multi-season \textit{NORDLAND} dataset, and the multi-weather \textit{GTAV} dataset. Experimental results show that our method improves the feature robustness against variant environmental conditions. : 6 pages, 5 figures, ICRA 2019 accepted Report Nordland Nordland Nordland DataCite Metadata Store (German National Library of Science and Technology) |
institution |
Open Polar |
collection |
DataCite Metadata Store (German National Library of Science and Technology) |
op_collection_id |
ftdatacite |
language |
unknown |
topic |
Robotics cs.RO Artificial Intelligence cs.AI Computer Vision and Pattern Recognition cs.CV FOS Computer and information sciences |
spellingShingle |
Robotics cs.RO Artificial Intelligence cs.AI Computer Vision and Pattern Recognition cs.CV FOS Computer and information sciences Yin, Peng Xu, Lingyun Li, Xueqian Yin, Chen Li, Yingli Srivatsan, Rangaprasad Arun Li, Lu Ji, Jianmin He, Yuqing A Multi-Domain Feature Learning Method for Visual Place Recognition |
topic_facet |
Robotics cs.RO Artificial Intelligence cs.AI Computer Vision and Pattern Recognition cs.CV FOS Computer and information sciences |
description |
Visual Place Recognition (VPR) is an important component in both computer vision and robotics applications, thanks to its ability to determine whether a place has been visited and where specifically. A major challenge in VPR is to handle changes of environmental conditions including weather, season and illumination. Most VPR methods try to improve the place recognition performance by ignoring the environmental factors, leading to decreased accuracy decreases when environmental conditions change significantly, such as day versus night. To this end, we propose an end-to-end conditional visual place recognition method. Specifically, we introduce the multi-domain feature learning method (MDFL) to capture multiple attribute-descriptions for a given place, and then use a feature detaching module to separate the environmental condition-related features from those that are not. The only label required within this feature learning pipeline is the environmental condition. Evaluation of the proposed method is conducted on the multi-season \textit{NORDLAND} dataset, and the multi-weather \textit{GTAV} dataset. Experimental results show that our method improves the feature robustness against variant environmental conditions. : 6 pages, 5 figures, ICRA 2019 accepted |
format |
Report |
author |
Yin, Peng Xu, Lingyun Li, Xueqian Yin, Chen Li, Yingli Srivatsan, Rangaprasad Arun Li, Lu Ji, Jianmin He, Yuqing |
author_facet |
Yin, Peng Xu, Lingyun Li, Xueqian Yin, Chen Li, Yingli Srivatsan, Rangaprasad Arun Li, Lu Ji, Jianmin He, Yuqing |
author_sort |
Yin, Peng |
title |
A Multi-Domain Feature Learning Method for Visual Place Recognition |
title_short |
A Multi-Domain Feature Learning Method for Visual Place Recognition |
title_full |
A Multi-Domain Feature Learning Method for Visual Place Recognition |
title_fullStr |
A Multi-Domain Feature Learning Method for Visual Place Recognition |
title_full_unstemmed |
A Multi-Domain Feature Learning Method for Visual Place Recognition |
title_sort |
multi-domain feature learning method for visual place recognition |
publisher |
arXiv |
publishDate |
2019 |
url |
https://dx.doi.org/10.48550/arxiv.1902.10058 https://arxiv.org/abs/1902.10058 |
genre |
Nordland Nordland Nordland |
genre_facet |
Nordland Nordland Nordland |
op_rights |
arXiv.org perpetual, non-exclusive license http://arxiv.org/licenses/nonexclusive-distrib/1.0/ |
op_doi |
https://doi.org/10.48550/arxiv.1902.10058 |
_version_ |
1766115750578225152 |