We provide step-by-step guides to localize with Aachen, InLoc, and to generate reference poses for your own data using SfM. Python -m hloc.name_of_script -arg1 -arg2 Tasks Hloc can be imported as an external package with import hloc or called from the command line with: hloc/pipelines/ : entire pipelines for multiple datasets.hloc/matchers/ : interfaces for feature matchers.hloc/extractors/ : interfaces for feature extractors.When 3D Lidar scans are available, such as for the indoor dataset InLoc, step 2. The localization can then be evaluated on for the supported datasets. Find database images relevant to each query, using retrieval.Triangulate a new SfM model with COLMAP.Match these database pairs with SuperGlue.Find covisible database images, with retrieval or a prior SfM model.Extract SuperPoint local features for all database and query images.The toolbox is composed of scripts, which roughly perform the following steps: Jupyter notebook -ip 0.0.0.0 -port 8888 -no-browser -allow-root General pipeline Installing the package locally pulls the other dependencies:ĭocker run -it -rm -p 8888:8888 hloc:latest # for GPU support, add `-runtime=nvidia`
Hloc requires Python >=3.7 and PyTorch >=1.1. Try it with your own data and let us know! Installation
The notebook demo.ipynb shows how to run SfM and localization in just a few steps. Hierachical Localization uses both image retrieval and feature matching Quick start ➡️īuild 3D maps with Structure-from-Motion and localize any Internet image right from your browser! You can now run hloc and COLMAP in Google Colab with GPU for free.
Implement new localization pipelines and debug them easily □.Evaluate your own local features or image retrieval for visual localization.Run Structure-from-Motion with SuperPoint+SuperGlue to localize with your own datasets.Reproduce our CVPR 2020 winning results on outdoor (Aachen) and indoor (InLoc) datasets.This codebase won the indoor/outdoor localization challenges at CVPR 2020 and ECCV 2020, in combination with SuperGlue, our graph neural network for feature matching. It implements Hierarchical Localization, leveraging image retrieval and feature matching, and is fast, accurate, and scalable. This is hloc, a modular toolbox for state-of-the-art 6-DoF visual localization. Hloc - the hierarchical localization toolbox