Dmytro Mishkin

caffenet-benchmark. Evaluation of the CNN design choices performance on ImageNet-2012.

pydegensac. Advanced RANSAC (DEGENSAC) with bells and whistles for H and F estimation

affnet. Code and weights for local feature affine shape estimation paper "Repeatability Is Not Enough: Learning Discriminative Affine Regions via Discriminability"

pytorch-sift. PyTorch implementation of SIFT descriptor

manifold-diffusion. Diffusion on manifolds for image retrieval

LSUVinit. Reference caffe implementation of LSUV initialization

pyransac. Fast and accurate python RANSAC with LO, LAF-check

mods. MODS (Matching On Demand with view Synthesis) is algorithm for wide-baseline matching.

ransac-tutorial-2020-data. Starter kit for the CVPR 2020 RANSAC tutorial benchmark

pymagsac. MAGSAC: marginalizing sample consensus, python version

LSUV-keras. Simple implementation of the LSUV initialization in keras

navigation-benchmark. Code for "Benchmarking Classic and Learned Navigation in Complex 3D Environments" paper

google-retrieval-challenge-2019-fastai-starter. fast.ai starter kit for Google Landmark Retrieval 2019 challenge

LSUV-pytorch. Simple implementation of the LSUV initialization in PyTorch

mods-light-zmq. MODS with external deep descriptors/detectors

whale-identification-2018. Solution to Whale Identification Challenge 2018

extract_patches. Function for local patch extraction from OpenCV keypoints with proper bluring

numpy-sift. Numpy implementation of SIFT descriptor

imc2023-kornia-starter-pack. Simple jupyter notebook for 3D reconstruction using kornia and pycolmap

kornia_moons. Conversions between kornia and other computer vision libraries formats

matching-strategies-comparison. Comparison of the matching strategies for local feature descriptor

fast_atan2. Realization of the atan2 approximations faster than standard function.

hesaff-pytorch. PyTorch implementation of Hessian-Affine local feature detector

caffe. Caffe: a Fast framework for deep learning. For the most recent version checkout the dev branch. For the latest stable release checkout the master branch.

wide-baseline-stereo-blog. Blog about wide baseline stereo and local features

hardnet-in-fastai2-and-kornia. Re-implementation of local descriptor HardNet training in fasta2+kornia

imc2021-sample-kornia-submission. Tutorial on how to create submission to Image Matching Challenge 2021

brown_phototour_revisited. New testing protocol for learning local patch descriptors on Brown Phototour dataset

pixelstitch. Simple tool for labelling the correspondences

cpp-extract-patches. C++ header-only lib for extracting local patches

extract-patches-old. Simple function for local patch extraction from OpenCV keypoints.

wxbs-descriptors-benchmark. W1BS local patch descriptors benchmark

mpv-templates-backup. Jupyter Notebook

vs3-cnn-labs. Computer vision labs for Vision and Sports Summer School 2022

wxbs-benchmark. Code for benchmarking image matchers on WxBS dataset/

caffe-preprocessing-scripts. Creates caffe lmdb from bunch of dirs with images. Clean-up, check, resize included

kornia. Open Source Differentiable Computer Vision Library

glue-factory. Training library for local feature detection and matching

DFE. Python

image-matching-benchmark. Public release of the Image Matching Benchmark: https://vision.uvic.ca/image-matching-challenge

LightGlue. LightGlue: Local Feature Matching at Light Speed

map-anything. MapAnything: Universal Feed-Forward Metric 3D Reconstruction