Conference item
Unsupervised place recognition with deep embedding learning over radar videos
- Abstract:
- We learn, in an unsupervised way, an embedding from sequences of radar images that is suitable for solving place recognition problem using complex radar data. We experiment on 280 km of data and show performance exceeding state-of-the-art supervised approaches, localising correctly 98.38 % of the time when using just the nearest database candidate.
- Publication status:
- Published
- Peer review status:
- Peer reviewed
Actions
Authors
- Publisher:
- IEEE
- Publication date:
- 2021-05-31
- Acceptance date:
- 2021-05-15
- Event title:
- 2021 ICRA Workshop: Radar Perception for All-Weather Autonomy
- Event location:
- Virtual event
- Event website:
- https://sites.google.com/view/radar-robotics/home
- Event start date:
- 2021-05-31
- Event end date:
- 2021-05-31
- Language:
-
English
- Keywords:
- Pubs id:
-
1242884
- Local pid:
-
pubs:1242884
- Deposit date:
-
2022-03-09
Terms of use
- Copyright date:
- 2021
- Notes:
- This paper was presented at the 2021 ICRA Workshop: Radar Perception for All-Weather Autonomy, 31 May 2021, Virtual event. This is the accepted manuscript version of the paper.
If you are the owner of this record, you can report an update to it here: Report update to this record