#213213th of 214 in ML frameworks and librariesFor image, video and design, 3d
hierarchical-localization
Python toolbox for 6-DoF visual localization and Structure-from-Motion
- Stars
- 4.3k
- License
- Apache-2.0
- Last commit
- Dec 2025
- Last release
- Jul 2023
- Language
- Python
Missed the latest GitHub check; these facts may be older.
Overview
hloc implements Hierarchical Localization: image retrieval picks candidate database images, local features are matched, and a camera pose is estimated against a COLMAP-built 3D SfM model. It wraps extractors (SuperPoint, DISK, D2-Net, SIFT, R2D2), matchers (SuperGlue, LightGlue, LoFTR, nearest neighbor) and retrieval models (NetVLAD, AP-GeM, OpenIBL, MegaLoc). Ready-made pipelines cover Aachen, InLoc, RobotCar, Cambridge Landmarks and 7-Scenes.
Who it is for: Computer vision researchers building visual localization or SfM pipelines
Strengths
- Pluggable extractor, matcher and retrieval interfaces via a BaseModel subclass
- Pipeline scripts for Aachen, InLoc, CMU Seasons, RobotCar, 4Seasons, Cambridge and 7-Scenes
- Since v1.3 pycolmap is enough; no separate COLMAP install
- Colab demo notebook runs SfM and localization on a free GPU
Weaknesses
- Last tagged release is v1.4 from July 2023
- Research toolbox of scripts and notebooks, not a packaged service or GUI
- Needs git submodules for external feature repos
- TensorFlow-based features require manual HDF5 export or code changes
What it needs
- GPU optional
- Docker
- Needs PyTorch >=1.1, pycolmap >=0.3.0, kornia >=0.6.11, Python >=3.7
- Models: SuperPoint, DISK, D2-Net, SIFT, R2D2
Also in ML frameworks and libraries
See all 214| Rank | Project | Score |
|---|---|---|
| 1 | transformersPython library defining open models for text, vision, audio and multimodal tasks | 94 out of 100 |
| 2 | ultralyticsPython library and CLI for training and running YOLO vision models | 94 out of 100 |
| 3 | pytorch-image-modelsPyTorch library of image models, pretrained weights and training scripts | 94 out of 100 |
| #4 | pytorchPython tensor library with GPU support and tape-based autograd | 93 out of 100 |
| #5 | supervisionPython toolkit for loading, annotating and counting computer vision detections | 92 out of 100 |