VisLoc

Abstract

VisLoc is a comprehensive collection of models and algorithms designed for use in Extended Reality (XR) and robotics applications, organized around three parts: retrieval, matching, and localization.

• Retrieval: algorithms and models for searching and retrieving visual data.
• Matching (ImMatch): a common interface for local feature extraction, image matching, and geometric estimation.
• Localization: image to 3D scene localization and absolute pose estimation.