Direct, Dense, and Deformable: Template-Based Non-Rigid 3D Reconstruction from RGB Video
- Submitting institution
-
University College London
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 14139
- Type
- E - Conference contribution
- DOI
-
10.1109/ICCV.2015.111
- Title of conference / published proceedings
- 2015 IEEE International Conference on Computer Vision (ICCV)
- First page
- 918
- Volume
- 2015
- Issue
- -
- ISSN
- 1550-5499
- Open access status
- Out of scope for open access requirements
- Month of publication
- December
- Year of publication
- 2015
- 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
- 13
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- First paper to propose a sequential photometric approach to track non-rigid 3D objects purely from video captured by a standard consumer camera, a long standing problem in computer vision. Previous methods needed mocap markers or depth/stereo inputs. The monocular tracking algorithms proposed in this paper constituted core technology that led to the foundation of Synthesia in 2017, a startup company pioneering AI powered video synthesis backed by $3.1M seed investment in early 2019 and with 17 employees as of late 2020, including 5 former PhD/MSc students from UCL.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -