Gespeichert in:
| Hauptverfasser: | , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2022
|
| Schlagworte: | |
| Online-Zugang: | https://arxiv.org/abs/2208.09315 |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866910705423220736 |
|---|---|
| author | Chen, Chao Cheng, Zegang Liu, Xinhao Li, Yiming Ding, Li Wang, Ruoyu Feng, Chen |
| author_facet | Chen, Chao Cheng, Zegang Liu, Xinhao Li, Yiming Ding, Li Wang, Ruoyu Feng, Chen |
| contents | Visual place recognition (VPR) using deep networks has achieved state-of-the-art performance. However, most of them require a training set with ground truth sensor poses to obtain positive and negative samples of each observation's spatial neighborhood for supervised learning. When such information is unavailable, temporal neighborhoods from a sequentially collected data stream could be exploited for self-supervised training, although we find its performance suboptimal. Inspired by noisy label learning, we propose a novel self-supervised framework named TF-VPR that uses temporal neighborhoods and learnable feature neighborhoods to discover unknown spatial neighborhoods. Our method follows an iterative training paradigm which alternates between: (1) representation learning with data augmentation, (2) positive set expansion to include the current feature space neighbors, and (3) positive set contraction via geometric verification. We conduct auto-labeling and generalization tests on both simulated and real datasets, with either RGB images or point clouds as inputs. The results show that our method outperforms self-supervised baselines in recall rate, robustness, and heading diversity, a novel metric we propose for VPR. Our code and datasets can be found at https://ai4ce.github.io/TF-VPR/ |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2208_09315 |
| institution | arXiv |
| publishDate | 2022 |
| record_format | arxiv |
| spellingShingle | Self-Supervised Place Recognition by Refining Temporal and Featural Pseudo Labels from Panoramic Data Chen, Chao Cheng, Zegang Liu, Xinhao Li, Yiming Ding, Li Wang, Ruoyu Feng, Chen Computer Vision and Pattern Recognition Visual place recognition (VPR) using deep networks has achieved state-of-the-art performance. However, most of them require a training set with ground truth sensor poses to obtain positive and negative samples of each observation's spatial neighborhood for supervised learning. When such information is unavailable, temporal neighborhoods from a sequentially collected data stream could be exploited for self-supervised training, although we find its performance suboptimal. Inspired by noisy label learning, we propose a novel self-supervised framework named TF-VPR that uses temporal neighborhoods and learnable feature neighborhoods to discover unknown spatial neighborhoods. Our method follows an iterative training paradigm which alternates between: (1) representation learning with data augmentation, (2) positive set expansion to include the current feature space neighbors, and (3) positive set contraction via geometric verification. We conduct auto-labeling and generalization tests on both simulated and real datasets, with either RGB images or point clouds as inputs. The results show that our method outperforms self-supervised baselines in recall rate, robustness, and heading diversity, a novel metric we propose for VPR. Our code and datasets can be found at https://ai4ce.github.io/TF-VPR/ |
| title | Self-Supervised Place Recognition by Refining Temporal and Featural Pseudo Labels from Panoramic Data |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2208.09315 |