What can Computer Vision learn from Ranganathan?
🤖 AI Summary
This work addresses the semantic gap between visual representations and lexical semantics in computer vision by proposing vTelos, a novel visual annotation methodology inspired by librarian S. R. Ranganathan’s faceted classification principles. For the first time, Ranganathan’s theory of faceted classification is introduced into computer vision to systematically construct a semantically aligned annotation framework, thereby enhancing both semantic consistency and structural coherence in datasets. Experimental results demonstrate that vTelos significantly improves annotation quality and boosts downstream model accuracy, effectively mitigating benchmark deficiencies caused by semantic inconsistency. The approach establishes a new theoretical foundation and practical pathway for constructing high-quality visual datasets with improved semantic fidelity.
Application Category
📝 Abstract
The Semantic Gap Problem (SGP) in Computer Vision (CV) arises from the misalignment between visual and lexical semantics leading to flawed CV dataset design and CV benchmarks. This paper proposes that classification principles of S.R. Ranganathan can offer a principled starting point to address SGP and design high-quality CV datasets. We elucidate how these principles, suitably adapted, underpin the vTelos CV annotation methodology. The paper also briefly presents experimental evidence showing improvements in CV annotation and accuracy, thereby, validating vTelos.
Problem
Research questions and friction points this paper is trying to address.
Semantic Gap Problem
Computer Vision
dataset design
lexical semantics
visual semantics
Innovation
Methods, ideas, or system contributions that make the work stand out.
Semantic Gap Problem
Ranganathan's classification principles
vTelos
computer vision annotation
dataset design
🔎 Similar Papers
No similar papers found.
