Awesome

Aggregated Selective Match Kernels (ASMK) for Image Retrieval

This is a Matlab package that we provide to reproduce the results of our ICCV 2013 paper. This code implements the ASMK* method, which offers the best trade-off between search accuracy and resource requirements (memory and speed). We additionally provide the code to reproduce the ASMK* results using DELF descriptors in our CVPR 2018 paper.

@InProceedings{TAJ13,
  author       = "Giorgos Tolias and Yannis Avrithis and Herv\'e J\'egou",
  title        = "To aggregate or not to aggregate: Selective match kernels for image search",
  booktitle    = "IEEE International Conference on Computer Vision",
  year         = "2013"
}

@InProceedings{RIT+18,
  author       = "Filip Radenovic, Ahmet Iscen, Giorgos Tolias, Yannis Avrithis, and Ondřej Chum",
  title        = "Revisiting Oxford and Paris: Large-Scale Image Retrieval Benchmarking",
  booktitle    = "IEEE Conference on Computer Vision and Patter Recognition ",
  year         = "2018"
}

Prerequisites

The prerequisites are automatically downloaded when running the main scripts.

Running (ICCV 2013)

To reproduce the experiments in our ICCV 2013 paper using Hessian Affine features and SIFT descriptors launch the test program in matlab:

>> test_asmk

Running (CVPR 2018)

To reproduce the experiments in our CVPR 2018 paper using DELF descriptors launch the following commands in matlab:

>> cd revisitop
>> setup
>> create_index
>> search_index