{"id":20842045,"url":"https://github.com/iago-suarez/efficient-descriptors","last_synced_at":"2025-05-08T22:41:07.654Z","repository":{"id":52149715,"uuid":"394737403","full_name":"iago-suarez/efficient-descriptors","owner":"iago-suarez","description":":rocket::rocket: Revisiting Binary Local Image Description for Resource Limited Devices","archived":false,"fork":false,"pushed_at":"2023-10-21T19:15:15.000Z","size":45876,"stargazers_count":170,"open_issues_count":3,"forks_count":27,"subscribers_count":8,"default_branch":"main","last_synced_at":"2025-05-02T16:03:33.694Z","etag":null,"topics":["computer-vision","descriptors","local-features","real-time","robotics","slam"],"latest_commit_sha":null,"homepage":"https://iago-suarez.com/efficient-descriptors","language":"C","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/iago-suarez.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null}},"created_at":"2021-08-10T17:56:38.000Z","updated_at":"2025-03-03T08:22:05.000Z","dependencies_parsed_at":"2022-09-18T05:02:09.172Z","dependency_job_id":null,"html_url":"https://github.com/iago-suarez/efficient-descriptors","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/iago-suarez%2Fefficient-descriptors","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/iago-suarez%2Fefficient-descriptors/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/iago-suarez%2Fefficient-descriptors/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/iago-suarez%2Fefficient-descriptors/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/iago-suarez","download_url":"https://codeload.github.com/iago-suarez/efficient-descriptors/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":253160727,"owners_count":21863624,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["computer-vision","descriptors","local-features","real-time","robotics","slam"],"created_at":"2024-11-18T01:22:52.235Z","updated_at":"2025-05-08T22:41:07.627Z","avatar_url":"https://github.com/iago-suarez.png","language":"C","funding_links":[],"categories":[],"sub_categories":[],"readme":"![Graffter Banner](imgs/banner.jpg)\n# Revisiting Binary Local Image Description for Resource Limited Devices\n\n**📢IMPORTANT📢 BAD is now available in OpenCV as [cv::xfeatures2d::TEBLID](https://docs.opencv.org/4.x/dd/dc1/classcv_1_1xfeatures2d_1_1TEBLID.html)**\n\nThis repository contains the source code of **BAD** (renamed as [TEBLID](https://docs.opencv.org/4.x/dd/dc1/classcv_1_1xfeatures2d_1_1TEBLID.html) after its inclusion in OpenCV contrib) and **HashSIFT** descriptors presented in \"[Revisiting Binary Local Image Description for Resource Limited Devices](https://arxiv.org/pdf/2108.08380.pdf)\". When accuracy and efficiency are both important, the descriptors in this repository offer the perfect trade-off for real-time applications and resource limited devices like smartphones, robots or drones. \n\n![Graffter header image](imgs/header.jpg)\n\n## Dependencies\nThe code depends on OpenCV 4.\n\u003cdetails\u003e \n\u003csummary\u003eTo install OpenCV ... \u003c/summary\u003e In Ubuntu 18.04 compile it from sources with the following instructions:\n\n```shell script\n# Install dependencies (Ubuntu 18.04)\nsudo apt-get install -y build-essential cmake git libgtk2.0-dev pkg-config libavcodec-dev libavformat-dev libswscale-dev\n# Download source code\ngit clone https://github.com/opencv/opencv.git --branch 4.5.2 --depth 1\n# Create build directory\ncd opencv \u0026\u0026 mkdir build \u0026\u0026 cd build\n# Generate makefiles, compile and install\ncmake -D CMAKE_BUILD_TYPE=RELEASE -D CMAKE_INSTALL_PREFIX=/usr/local ..\nmake -j\nsudo make install\n```\n\n* NOTE: The code also compile with OpenCV 3, but without parallel for's. Thus, the description will be slower.\n\u003c/details\u003e\n\n## Compile and Run\nWe provide the pre-trained execution code of BAD and HashSIFT descriptors. \nThe code contains two demos. The first one estimates the fundamental matrix between \ntwo images of the Machine Hall 05 from [EuRoC MAV Dataset](https://projects.asl.ethz.ch/datasets/doku.php?id=kmavvisualinertialdatasets).\nThe demo detects feature points using ORB detector (FAST + Harris score) and describes using BAD.\nWe draw the RANSAC inliers with less than 3px of epipolar error.\nThe code can be compiled with Cmake:\n\n```shell script\nmkdir build \u0026\u0026 cd build\ncmake .. \u0026\u0026 make\n./stereo_demo [hashsift]\n```\n\n![](imgs/stereo-euroc-compressed.gif)\n\nWe also show a second demo that registers a pair of planar images.\n\n```shell script\n./homography_demo [hashsift]\n```\n\nThe result for the provided images should be several imshows and something like this in the standard output:\n```\n*************** Homography estimation demo ***************\nDetected features: 1000\nMatched features:   64\nInliers percentage:  6.4%\n```\nIf ORB descriptor is used instead of BAD, only 2.4% of inliers are obtained.\n\n## References\nIf you use this code, you must cite our Robotics and Automation Letters paper:\n```bibtex\n@ARTICLE{9521740,\n  author={Su\\'arez, Iago and Buenaposada, Jos\\'e M. and Baumela, Luis},\n  journal={IEEE Robotics and Automation Letters}, \n  title={Revisiting Binary Local Image Description for Resource Limited Devices}, \n  year={2021},\n  volume={},\n  number={},\n  pages={1-1},\n  doi={10.1109/LRA.2021.3107024}}\n```\n\n## Contact and Licence\nWe provide a free pre-trained version of the execution code. Full execution and training code can be obtained under license, if you are interested please contact us:\n\n* Iago Suárez ( iago.suarez@thegraffter.com ) for technical issues.\n* Miguel Ángel Orellana Sainz ( miguel.orellana@thegraffter.com ) for commercial issues.\n\nThis software was developed by [The Graffter S.L.](http://www.thegraffter.com) in collaboration with the [PCR lab of the Universidad Politécnica de Madrid](http://www.dia.fi.upm.es/~pcr/research.html).\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fiago-suarez%2Fefficient-descriptors","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fiago-suarez%2Fefficient-descriptors","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fiago-suarez%2Fefficient-descriptors/lists"}