CBIR using Pre-Trained Neural Networks

TL;DR

This paper proposes a novel image retrieval method using a pretrained Inception V3 model with multi-branch feature extraction and combination, achieving high accuracy on the CUB200-2011 dataset.

Contribution

It introduces a multi-branch feature extraction approach using a pretrained neural network for improved image retrieval accuracy.

Findings

Achieved 99.46% training accuracy on CUB200-2011

Improved validation accuracy to 88.89% with multi-branch features

Utilized MS-RMAC for enhanced feature extraction

Abstract

Much of the recent research work in image retrieval, has been focused around using Neural Networks as the core component. Many of the papers in other domain have shown that training multiple models, and then combining their outcomes, provide good results. This is since, a single Neural Network model, may not extract sufficient information from the input. In this paper, we aim to follow a different approach. Instead of the using a single model, we use a pretrained Inception V3 model, and extract activation of its last fully connected layer, which forms a low dimensional representation of the image. This feature matrix, is then divided into branches and separate feature extraction is done for each branch, to obtain multiple features flattened into a vector. Such individual vectors are then combined, to get a single combined feature. We make use of CUB200-2011 Dataset, which comprises of 200 birds classes to train the model on. We achieved a training accuracy of 99.46% and validation accuracy of 84.56% for the same. On further use of 3 branched global descriptors, we improve the validation accuracy to 88.89%. For this, we made use of MS-RMAC feature extraction method.