Web image search engine based on LSH index and CNN Resnet50
This work addresses efficiency in web image search engines for large-scale data, but it is incremental as it combines existing LSH and CNN methods without introducing new paradigms.
The paper tackled the problem of slow exact search in Content-Based Image Retrieval (CBIR) systems for large image collections by implementing a system using Locality Sensitive Hashing (LSH) indexing with CNN features from ResNet50 and ResNet50v2, achieving performance analysis with mAP values to evaluate speed and accuracy trade-offs.
To implement a good Content Based Image Retrieval (CBIR) system, it is essential to adopt efficient search methods. One way to achieve this results is by exploiting approximate search techniques. In fact, when we deal with very large collections of data, using an exact search method makes the system very slow. In this project, we adopt the Locality Sensitive Hashing (LSH) index to implement a CBIR system that allows us to perform fast similarity search on deep features. Specifically, we exploit transfer learning techniques to extract deep features from images; this phase is done using two famous Convolutional Neural Networks (CNNs) as features extractors: Resnet50 and Resnet50v2, both pre-trained on ImageNet. Then we try out several fully connected deep neural networks, built on top of both of the previously mentioned CNNs in order to fine-tuned them on our dataset. In both of previous cases, we index the features within our LSH index implementation and within a sequential scan, to better understand how much the introduction of the index affects the results. Finally, we carry out a performance analysis: we evaluate the relevance of the result set, computing the mAP (mean Average Precision) value obtained during the different experiments with respect to the number of done comparison and varying the hyper-parameter values of the LSH index.