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

Born-Infeld (BI) for AI: Energy-Conserving Descent (ECD) for Optimization

A minimalist environment for decision-making in autonomous driving

Group Activity Recognition with Clustered Spatial Temporal Transformer

RobustVideoMatting and background composing in one model by using onnxruntime.

SAT: 2D Semantics Assisted Training for 3D Visual Grounding, ICCV 2021 (Oral)

implementation of paper - You Only Learn One Representation: Unified Network for Multiple Tasks

[NeurIPS-2021] Slow Learning and Fast Inference: Efficient Graph Similarity Computation via Knowledge Distillation

Elegy is a framework-agnostic Trainer interface for the Jax ecosystem.

Implementation of Cross Transformer for spatially-aware few-shot transfer, in Pytorch

Graph neural network message passing reframed as a Transformer with local attention

A code repository associated with the paper A Benchmark for Rough Sketch Cleanup by Chuan Yan, David Vanderhaeghe, and Yotam Gingold from SIGGRAPH Asia 2020.

Angle data is a simple data type.

This is an official implementation for "AS-MLP: An Axial Shifted MLP Architecture for Vision".

AdelaiDepth is an open source toolbox for monocular depth prediction.

Easy-to-use library to boost AI inference leveraging state-of-the-art optimization techniques.

[ArXiv 2021] Data-Efficient Instance Generation from Instance Discrimination

a reimplementation of LiteFlowNet in PyTorch that matches the official Caffe version

My 1st place solution at Kaggle Hotel-ID 2021

The Pytorch code of "Joint Distribution Matters: Deep Brownian Distance Covariance for Few-Shot Classification", CVPR 2022 (Oral).

This is the reference implementation for "Coresets via Bilevel Optimization for Continual Learning and Streaming"