A benchmark for RGB-D visual odometry, 3D reconstruction and SLAM
- Submitting institution
-
Imperial College of Science, Technology and Medicine
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 2131
- Type
- E - Conference contribution
- DOI
-
10.1109/ICRA.2014.6907054
- Title of conference / published proceedings
- 2014 IEEE International Conference on Robotics and Automation (ICRA)
- First page
- 1524
- Volume
- -
- Issue
- -
- ISSN
- 1050-4729
- Open access status
- Out of scope for open access requirements
- Month of publication
- September
- Year of publication
- 2014
- URL
-
-
- Supplementary information
-
-
- 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
- 209
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- This collaboration with NUIM Ireland presents a breakthrough new type of SLAM evaluation framework, which is highly realistic but fully synthetic, meaning that it supports controlled experiments with full ground truth on scene shape. Many research projects in academia and industry (e.g. BundleFusion from Stanford/MPI/Microsoft; http://graphics.stanford.edu/projects/bundlefusion and DSO from TUM/Intel; https://vision.in.tum.de/research/vslam/dso) have used our evaluation framework, and it has inspired significant new projects (e.g. Stanford's ScanNet; http://www.scan-net.org/, Intel's Playing for Data; https://link.springer.com/chapter/10.1007/978-3-319-46475-6_7, PAMELA's SLAMBench; http://apt.cs.manchester.ac.uk/projects/PAMELA/tools/SLAMBench/ and the Dyson Robotics Lab's SceneNet-RGBD; https://www.imperial.ac.uk/dyson-robotics-lab/downloads/scenenet-rgb-d-software/).
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -
