ARCHIVES
VOL. 1, ISSUE 1 (2025)
Advancements in content-based image retrieval: A comprehensive survey of relevance feedback techniques
Authors
Moumita Barik
Abstract
Content-Based Image Retrieval (CBIR) has emerged as
a fundamental technology for managing and searching large-scale image databases
by analyzing visual content rather than relying on textual annotations.
However, the semantic gap between low-level visual features and high-level
human perception remains a significant challenge in achieving accurate retrieval
results. Relevance feedback techniques have proven to be one of the most
effective approaches to bridge this semantic gap by incorporating user
interactions to refine search queries iteratively. This comprehensive survey
examines the evolution and current state of relevance feedback methods in CBIR
systems, categorizing approaches into positive feedback, negative feedback, and
hybrid techniques. We analyze traditional machine learning approaches including
Support Vector Machines, Neural Networks, and clustering-based methods,
alongside recent deep learning innovations such as Convolutional Neural
Networks and attention mechanisms. The survey also covers evaluation
methodologies, benchmark datasets, and emerging trends including active
learning, multi-modal feedback, and real-time adaptation. Our analysis reveals
that while significant progress has been made, challenges persist in handling
concept drift, scalability, and user experience optimization. This work
provides researchers and practitioners with a thorough understanding of the
current landscape and identifies promising directions for future research in
relevance feedback for CBIR systems.
Download
Pages:5-9
How to cite this article:
Moumita Barik "Advancements in content-based image retrieval: A comprehensive survey of relevance feedback techniques". World Journal of Food Science, Vol 1, Issue 1, 2025, Pages 5-9
Download Author Certificate
Please enter the email address corresponding to this article submission to download your certificate.