NIST: An Image Classification Network to Image Semantic Retrieval

TL;DR

This paper introduces NIST, a novel image semantic retrieval framework utilizing a classification network to improve retrieval accuracy and efficiency by leveraging semantic features and a fusion strategy, achieving state-of-the-art results.

Contribution

The paper presents a new semantic retrieval method based on GoogleNet features and a semantic distance algorithm, reducing storage and computation compared to traditional feature matching.

Findings

Achieves state-of-the-art retrieval performance on MIRFLICKR-25K dataset.

Reduces storage and computation through semantic feature fusion.

Outperforms traditional feature matching methods.

Abstract

This paper proposes a classification network to image semantic retrieval (NIST) framework to counter the image retrieval challenge. Our approach leverages the successful classification network GoogleNet based on Convolutional Neural Networks to obtain the semantic feature matrix which contains the serial number of classes and corresponding probabilities. Compared with traditional image retrieval using feature matching to compute the similarity between two images, NIST leverages the semantic information to construct semantic feature matrix and uses the semantic distance algorithm to compute the similarity. Besides, the fusion strategy can significantly reduce storage and time consumption due to less classes participating in the last semantic distance computation. Experiments demonstrate that our NIST framework produces state-of-the-art results in retrieval experiments on MIRFLICKR-25K dataset.