End-to-end, sequence-to-sequence probabilistic visual odometry through deep neural networks
- Submitting institution
-
Imperial College of Science, Technology and Medicine
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 2406
- Type
- D - Journal article
- DOI
-
10.1177/0278364917734298
- Title of journal
- International Journal of Robotics Research
- Article number
- -
- First page
- 513
- Volume
- 37
- Issue
- 4-5
- ISSN
- 0278-3649
- Open access status
- Exception within 3 months of publication
- Month of publication
- October
- Year of publication
- 2017
- URL
-
-
- Supplementary information
-
10.1177/0278364917734298
- 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
-
3
- Research group(s)
-
-
- Citation count
- 35
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- Our CNN-RNN model pioneers a sequence learning approach to visual odometry and enables robust tracking of the motion of a monocular camera in real-time. This IJRR paper is an extension of a highly-cited ICRA'17 paper (https://doi.org/10.1109/ICRA.2017.7989236). The algorithm was patented (AU2018208816) and licensed by Oxford University spinout Navenio (https://www.navenio.com; contact: FoEREF@ic.ac.uk).
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -
