PyTorch implementation of SIFT descriptor

Last update: Dec 24, 2022

Overview

This is an differentiable pytorch implementation of SIFT patch descriptor. It is very slow for describing one patch, but quite fast for batch. It can be used for descriptop-based learning shape of affine feature.

UPD 08/2019 : pytorch-sift is added to kornia and available by kornia.features.SIFTDescriptor

There are different implementations of the SIFT on the web. I tried to match Michal Perdoch implementation, which gives high quality features for image retrieval CVPR2009. However, on planar datasets, it is inferior to vlfeat implementation. The main difference is gaussian weighting window parameters, so I have made a vlfeat-like version too. MP version weights patch center much more (see image below, left) and additionally crops everything outside the circular region. Right is vlfeat version

descriptor_mp_mode = SIFTNet(patch_size = 65,
                        sigma_type= 'hesamp',
                        masktype='CircularGauss')

descriptor_vlfeat_mode = SIFTNet(patch_size = 65,
                        sigma_type= 'vlfeat',
                        masktype='Gauss')

Results:

OPENCV-SIFT - mAP 
   Easy     Hard      Tough     mean
-------  -------  ---------  -------
0.47788  0.20997  0.0967711  0.26154

VLFeat-SIFT - mAP 
    Easy      Hard      Tough      mean
--------  --------  ---------  --------
0.466584  0.203966  0.0935743  0.254708

PYTORCH-SIFT-VLFEAT-65 - mAP 
    Easy      Hard      Tough      mean
--------  --------  ---------  --------
0.472563  0.202458  0.0910371  0.255353

NUMPY-SIFT-VLFEAT-65 - mAP 
    Easy      Hard      Tough      mean
--------  --------  ---------  --------
0.449431  0.197918  0.0905395  0.245963

PYTORCH-SIFT-MP-65 - mAP 
    Easy      Hard      Tough      mean
--------  --------  ---------  --------
0.430887  0.184834  0.0832707  0.232997

NUMPY-SIFT-MP-65 - mAP 
    Easy     Hard      Tough      mean
--------  -------  ---------  --------
0.417296  0.18114  0.0820582  0.226832

Speed:

0.00246 s per 65x65 patch - numpy SIFT
0.00028 s per 65x65 patch - C++ SIFT
0.00074 s per 65x65 patch - CPU, 256 patches per batch
0.00038 s per 65x65 patch - GPU (GM940, mobile), 256 patches per batch
0.00038 s per 65x65 patch - GPU (GM940, mobile), 256 patches per batch

If you use this code for academic purposes, please cite the following paper:

@InProceedings{AffNet2018,
    title = {Repeatability Is Not Enough: Learning Affine Regions via Discriminability},
    author = {Dmytro Mishkin, Filip Radenovic, Jiri Matas},
    booktitle = {Proceedings of ECCV},
    year = 2018,
    month = sep
}

PyTorch implementation of SIFT descriptor

Related tags

Overview

Owner

Dmytro Mishkin

Real-Time-Student-Attendence-System - Real Time Student Attendence System

Kaggle competition: Springleaf Marketing Response

Python Environment for Bayesian Learning

Safe Bayesian Optimization

Real-ESRGAN: Training Real-World Blind Super-Resolution with Pure Synthetic Data

RoMA: Robust Model Adaptation for Offline Model-based Optimization

An example project demonstrating how the Autonomous Learning Library can be used to build new reinforcement learning agents.

Temporally Coherent GAN SIGGRAPH project.

Code to accompany the paper "Finding Bipartite Components in Hypergraphs", which is published in NeurIPS'21.

PyTorch implementation for SDEdit: Image Synthesis and Editing with Stochastic Differential Equations

Hso-groupie - A pwnable challenge in Real World CTF 4th

Ludwig Benchmarking Toolkit

A texturizer that I just made. Nothing special here.

Compare neural networks by their feature similarity

Applying curriculum to meta-learning for few shot classification

Semi-supervised Stance Detection of Tweets Via Distant Network Supervision

A unified 3D Transformer Pipeline for visual synthesis

CLEAR algorithm for multi-view data association

AI assistant built in python.the features are it can display time,say weather,open-google,youtube,instagram.

ktrain is a Python library that makes deep learning and AI more accessible and easier to apply