BigSUR: Large-scale Structured Urban Reconstruction
- Submitting institution
-
The University of Leeds
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- UOA11-3862
- Type
- D - Journal article
- DOI
-
10.1145/3130800.3130823
- Title of journal
- ACM Transactions on Graphics
- Article number
- 204
- First page
- -
- Volume
- 36
- Issue
- 6
- ISSN
- 0730-0301
- Open access status
- Technical exception
- Month of publication
- November
- Year of publication
- 2017
- 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
- Yes
- Number of additional authors
-
3
- Research group(s)
-
D - CSE (Computational Science and Engineering)
- Citation count
- 19
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- Accurately modeling our built environment is critical to many goals from self-driving cars to city planning. However, many data sources (maps, photos..) are noisy with different properties and dimensionalities. BigSUR introduces a novel algorithm to, for the first time, fuse this multitude of sources into clean, semantically-labelled, 3D meshes. These architectural-quality models balance errors and remove inconsistency present in the input data. This was scaled to kilometer sized outputs encompassing 1,011 buildings and 37 city blocks from central London. BigSUR was instrumental to a Google Research Award (https://bit.ly/2XmIbiH). 4 minute Video (https://youtu.be/_lHRTBkC-yo).
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -
