Back to Search
Start Over
Zero-Shot Recognition Enhancement by Distance-Weighted Contextual Inference
- Source :
- Applied Sciences, Volume 10, Issue 20, Applied Sciences, Vol 10, Iss 7234, p 7234 (2020)
- Publication Year :
- 2020
- Publisher :
- MDPI AG, 2020.
-
Abstract
- Zero-shot recognition (ZSR) aims to perform visual classification by category in the absence of training samples. The focus in most traditional ZSR models is using semantic knowledge about familiar categories to represent unfamiliar categories with only the visual appearance of an unseen object. In this research, we consider not only visual information but context to enhance the classifier&rsquo<br />s cognitive ability in a multi-object scene. We propose a novel method, contextual inference, that uses external resources such as knowledge graphs and semantic embedding spaces to obtain similarity measures between an unseen object and its surrounding objects. Using the intuition that close contexts involve more related associations than distant ones, distance weighting is applied to each piece of surrounding information with a newly defined distance calculation formula. We integrated contextual inference into traditional ZSR models to calibrate their visual predictions, and performed extensive experiments on two different datasets for comparative evaluations. The experimental results demonstrate the effectiveness of our method through significant enhancements in performance.
- Subjects :
- 0209 industrial biotechnology
Computer science
Inference
Context (language use)
02 engineering and technology
lcsh:Technology
lcsh:Chemistry
distance-weighting
020901 industrial engineering & automation
Similarity (psychology)
0202 electrical engineering, electronic engineering, information engineering
Semantic memory
General Materials Science
lcsh:QH301-705.5
Instrumentation
Fluid Flow and Transfer Processes
lcsh:T
business.industry
zero-shot recognition
Process Chemistry and Technology
General Engineering
Pattern recognition
similarity measures
Object (computer science)
Visual appearance
lcsh:QC1-999
Computer Science Applications
Weighting
lcsh:Biology (General)
lcsh:QD1-999
knowledge graph
lcsh:TA1-2040
020201 artificial intelligence & image processing
Artificial intelligence
lcsh:Engineering (General). Civil engineering (General)
semantic embedding
business
Classifier (UML)
lcsh:Physics
Subjects
Details
- ISSN :
- 20763417
- Volume :
- 10
- Database :
- OpenAIRE
- Journal :
- Applied Sciences
- Accession number :
- edsair.doi.dedup.....485074795685ecda047e38532a8e1c88