Loop Closure Detection in RGB-D SLAM by Utilizing Siamese ConvNet Features
Loop closure detection is a key challenge in visual simultaneous localization and mapping (SLAM) systems, which has attracted significant research interest in recent years. It entails correctly determining whether a scene has previously been visited by a mobile robot and completely establishing the...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-12-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/12/1/62 |
| _version_ | 1797499707888500736 |
|---|---|
| author | Gang Xu Xiang Li Xingyu Zhang Guangxin Xing Feng Pan |
| author_facet | Gang Xu Xiang Li Xingyu Zhang Guangxin Xing Feng Pan |
| author_sort | Gang Xu |
| collection | DOAJ |
| description | Loop closure detection is a key challenge in visual simultaneous localization and mapping (SLAM) systems, which has attracted significant research interest in recent years. It entails correctly determining whether a scene has previously been visited by a mobile robot and completely establishing the consistent maps of motion. There are many loop closure detection methods that have been proposed, but most of these algorithms are handcrafted features-based and perform weak robustness to illumination variations. In this paper, we investigate a Siamese Convolutional Neural Network (SCNN) to solve the task of loop closure detection in RGB-D SLAM. Firstly, we use a pre-trained SCNN model to extract features as image descriptors; then, the L2 norm distance is adopted as a similarity metric between descriptors. In terms of the learned features for matching, there are two key issues for discussion: (1) how to define an appropriate loss as supervision (utilizing the cross-entropy loss, the contrastive loss, or the combination of two); and (2) how to combine the appearance information in RGB images and position information in depth images (utilizing early fusion, mid-level fusion or late fusion). We compare our proposed method of different baseline by experiments carried out on two public datasets (New College and NYU), and our performance outperforms the state-of-the-art. |
| first_indexed | 2024-03-10T03:51:18Z |
| format | Article |
| id | doaj.art-74990f96cc5f41829575d5542c719015 |
| institution | Directory Open Access Journal |
| issn | 2076-3417 |
| language | English |
| last_indexed | 2024-03-10T03:51:18Z |
| publishDate | 2021-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj.art-74990f96cc5f41829575d5542c7190152023-11-23T11:07:02ZengMDPI AGApplied Sciences2076-34172021-12-011216210.3390/app12010062Loop Closure Detection in RGB-D SLAM by Utilizing Siamese ConvNet FeaturesGang Xu0Xiang Li1Xingyu Zhang2Guangxin Xing3Feng Pan4Key Laboratory of Advanced Process Control for Light Industry (Ministry of Education), Jiangnan University, Wuxi 214000, ChinaAnhui Key Laboratory of Detection Technology and Energy Saving Devices, Anhui Polytechnic University, Wuhu 241000, ChinaAnhui Key Laboratory of Detection Technology and Energy Saving Devices, Anhui Polytechnic University, Wuhu 241000, ChinaAnhui Key Laboratory of Detection Technology and Energy Saving Devices, Anhui Polytechnic University, Wuhu 241000, ChinaKey Laboratory of Advanced Process Control for Light Industry (Ministry of Education), Jiangnan University, Wuxi 214000, ChinaLoop closure detection is a key challenge in visual simultaneous localization and mapping (SLAM) systems, which has attracted significant research interest in recent years. It entails correctly determining whether a scene has previously been visited by a mobile robot and completely establishing the consistent maps of motion. There are many loop closure detection methods that have been proposed, but most of these algorithms are handcrafted features-based and perform weak robustness to illumination variations. In this paper, we investigate a Siamese Convolutional Neural Network (SCNN) to solve the task of loop closure detection in RGB-D SLAM. Firstly, we use a pre-trained SCNN model to extract features as image descriptors; then, the L2 norm distance is adopted as a similarity metric between descriptors. In terms of the learned features for matching, there are two key issues for discussion: (1) how to define an appropriate loss as supervision (utilizing the cross-entropy loss, the contrastive loss, or the combination of two); and (2) how to combine the appearance information in RGB images and position information in depth images (utilizing early fusion, mid-level fusion or late fusion). We compare our proposed method of different baseline by experiments carried out on two public datasets (New College and NYU), and our performance outperforms the state-of-the-art.https://www.mdpi.com/2076-3417/12/1/62Siamese convolutional neural networkSLAMloop closure detectionRGB-D |
| spellingShingle | Gang Xu Xiang Li Xingyu Zhang Guangxin Xing Feng Pan Loop Closure Detection in RGB-D SLAM by Utilizing Siamese ConvNet Features Applied Sciences Siamese convolutional neural network SLAM loop closure detection RGB-D |
| title | Loop Closure Detection in RGB-D SLAM by Utilizing Siamese ConvNet Features |
| title_full | Loop Closure Detection in RGB-D SLAM by Utilizing Siamese ConvNet Features |
| title_fullStr | Loop Closure Detection in RGB-D SLAM by Utilizing Siamese ConvNet Features |
| title_full_unstemmed | Loop Closure Detection in RGB-D SLAM by Utilizing Siamese ConvNet Features |
| title_short | Loop Closure Detection in RGB-D SLAM by Utilizing Siamese ConvNet Features |
| title_sort | loop closure detection in rgb d slam by utilizing siamese convnet features |
| topic | Siamese convolutional neural network SLAM loop closure detection RGB-D |
| url | https://www.mdpi.com/2076-3417/12/1/62 |
| work_keys_str_mv | AT gangxu loopclosuredetectioninrgbdslambyutilizingsiameseconvnetfeatures AT xiangli loopclosuredetectioninrgbdslambyutilizingsiameseconvnetfeatures AT xingyuzhang loopclosuredetectioninrgbdslambyutilizingsiameseconvnetfeatures AT guangxinxing loopclosuredetectioninrgbdslambyutilizingsiameseconvnetfeatures AT fengpan loopclosuredetectioninrgbdslambyutilizingsiameseconvnetfeatures |