CodeSLAM - Learning a Compact, Optimisable Representation for Dense Visual SLAM
- Submitting institution
-
Imperial College of Science, Technology and Medicine
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 2210
- Type
- E - Conference contribution
- DOI
-
10.1109/CVPR.2018.00271
- Title of conference / published proceedings
- 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition
- First page
- 2560
- Volume
- -
- Issue
- -
- ISSN
- 1063-6919
- Open access status
- Compliant
- Month of publication
- December
- Year of publication
- 2018
- URL
-
-
- Supplementary information
-
10.1109/CVPR.2018.00271
- Request cross-referral to
- -
- Output has been delayed by COVID-19
- No
- COVID-19 affected output statement
- -
- Forensic science
- No
- Criminology
- No
- Interdisciplinary
- No
- Number of additional authors
-
4
- Research group(s)
-
-
- Citation count
- 22
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- This work follows a direct approach to bring deep learning into spatial scene understanding, which can fundamentally change the way a map is represented and stored. Our £4.5M EPSRC/Dyson Prosperity Partnership project (EP/SO36636/1) to expand the Dyson Robotics Lab has methods building from this paper at its core. This paper was accepted for oral presentation at CVPR 2018 (acceptance rate: 2%) and received a Best Paper Award Honourable Mention, rating it among the top 5 papers out of over 3300 submissions.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -
