Zero-Shot Recognition Enhancement by Distance-Weighted Contextual Inference
- Title
- Zero-Shot Recognition Enhancement by Distance-Weighted Contextual Inference
- Author
- 최용석
- Keywords
- zero-shot recognition; similarity measures; distance-weighting; knowledge graph; semantic embedding
- Issue Date
- 2020-10
- Publisher
- MDPI
- Citation
- APPLIED SCIENCES-BASEL, v. 10, no. 20, article no. 7234
- 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'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.
- URI
- https://www.mdpi.com/2076-3417/10/20/7234https://repository.hanyang.ac.kr/handle/20.500.11754/171639
- ISSN
- 2076-3417
- DOI
- 10.3390/app10207234
- Appears in Collections:
- COLLEGE OF ENGINEERING[S](공과대학) > COMPUTER SCIENCE(컴퓨터소프트웨어학부) > Articles
- Files in This Item:
There are no files associated with this item.
- Export
- RIS (EndNote)
- XLS (Excel)
- XML