search

SathwikTejaswi / deep-ranking

Learning Fine-grained Image Similarity with Deep Ranking is a novel application of neural networks, where the authors use a new multi scale architecture combined with a triplet loss to create a neural network that is able to perform image search. This repository is a simplified implementation of the same
MIT License
81 stars 17 forks source link
convolutional-neural-networks deep-learning deep-neural-networks image image-classification image-rank image-retrieval image-search neural-network python-3 python3 pytorch resnet resnet-50 tiny-imagenet200 triplet-loss

readme

Deep-Image-Ranking

Neural Networks have been used for a variety of tasks, especially using unstructured data. Neural Networks are extremely good at image recognition, image segmentation etc. Learning Fine-grained Image Similarity with Deep Ranking (https://users.eecs.northwestern.edu/~jwa368/pdfs/deep_ranking.pdf) is a novel application of neural networks, where the authors use a new multi scale architecture combined with a triplet loss to create a neural network that is able to perform image search.

This repository is a simpler implementation of the paper. The differences is that, the entire multi scale network has been replaced by a resnet. A simpler version of triplet sampling has been used.

Specifics :

Sample output

Sample results from the network are as shown below :

Query Image :

im1

Results :

im2 im2 im2 im2 im2