ResNet-101 Feature Map Extraction and Similarity Matching
Overview:
This repository contains a program that utilizes the powerful ResNet-101 convolutional neural network to extract feature maps from images, transforms them into vector representations, and loads them into Qdrant – a vector search engine for efficient similarity matching. The program is designed to find and return the top 10 most similar images to a given input image based on their feature map proximity in the vector space.