Baseline fusion for image an pattern recognition - what not to do (and how to do better)
- Submitting institution
-
University of St Andrews
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 251887717
- Type
- D - Journal article
- DOI
-
10.3390/jimaging3040044
- Title of journal
- Journal of Imaging
- Article number
- 44
- First page
- 1
- Volume
- 3
- Issue
- 4
- ISSN
- 2313-433X
- Open access status
- Compliant
- Month of publication
- October
- 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
- No
- Number of additional authors
-
0
- Research group(s)
-
A - Artificial Intelligence
- Citation count
- 2
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- In this work it is demonstrated that possibly the most commonly used baseline against which novel fusion methods are judged is fundamentally flawed, thus raising serious doubts about the performance and usefulness of numerous algorithms in the literature. This contribution is extremely important considering that multi-modal, fusion based decision making is pervasive in modern artificial intelligence and machine learning.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -
