Learning to Observe: Approximating Human Perceptual Thresholds for Detection of Suprathreshold Image Transformations
- Submitting institution
-
Birmingham City University
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 11Z_OP_E2005
- Type
- E - Conference contribution
- DOI
-
10.1109/CVPR42600.2020.00485
- Title of conference / published proceedings
- 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
- First page
- 4797
- Volume
- -
- Issue
- -
- ISSN
- -
- Open access status
- -
- Month of publication
- June
- Year of publication
- 2020
- 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
-
-
- Research group(s)
-
-
- Citation count
- -
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- This paper presents a method of approximation of perceptual functions based on human observers’ data. The dataset is available at https://cove.thecvf.com/datasets/333. Results from this work have been used to inform further work on time-based deep learning performance analysis (DOI 10.1109/DLS51937.2020.00007).
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -