1 个仓库
Integrated systems that combine image retrieval and pose estimation to locate cameras in 3D space.
Distinct from 3D Pose Estimation: Distinct from general pose estimation by incorporating a full pipeline of retrieval and matching against a reference model.
Explore 1 awesome GitHub repository matching artificial intelligence & ml · Visual Localization Frameworks. Refine with filters or upvote what's useful.
This project is a 3D visual localization framework designed to determine a camera's exact position and orientation by matching 2D image features against a 3D reference model. It includes a structure-from-motion pipeline to reconstruct 3D scene geometry from unordered image sets, creating the necessary spatial maps for localization. The system employs a hierarchical coarse-to-fine localization approach. This process begins with a global-descriptor image retrieval system to identify candidate reference images from a large database and progresses through local feature matching to final 3D model
Provides a complete system for determining camera poses by matching 2D features against 3D reference models.