Gravity aligned wireframes improve indoor camera localization without depth data
GALoc: Gravity Aligned Wireframes for Depth-Free Monocular Floorplan Localization
Computer Vision and Pattern RecognitionRobotics
Summary
Indoor navigation often uses floorplans to help figure out where a camera is, but current methods usually depend on estimating how far things are using depth data, which can fail in messy scenes. The authors propose GALoc, a method that skips depth estimation and instead uses geometric clues aligned with gravity to build wireframe maps of walls and floors. By using camera and sensor data, GALoc matches the camera's view to the floorplan more reliably, especially when enough wall structure is visible. Their tests showed that GALoc works as well or better than methods relying on depth, and it avoids errors when the scene lacks strong structure.
floorplan localizationmonocular RGB cameragravity alignmentwireframe modelingindoor mappingverticality constraintscoplanarityIMU orientationSE(2) searchindoor navigation
Authors
Jeahn Han, Minji Kim, Jeongbin Sohn, Jonghyeok Park, Matthias Wuest, Pyojin Kim
Abstract
Floorplans are compact, appearance-invariant maps ideal for indoor localization, yet existing methods rely on depth networks that are brittle in cluttered scenes. We propose GALoc, a geometry-first framework that replaces depth prediction with gravity-aligned wireframes that satisfy verticality and coplanarity by construction. Given monocular RGB, camera intrinsics, relative poses, and IMU orientation, GALoc constructs a linear constraint matrix encoding verticality and coplanarity, and finds the camera gauge minimizing its smallest singular value via global search. The rectified wireframes are projected into bird's-eye-view layouts through a closed-form, FOV-consistent transformation and matched against the floorplan via metric-free SE(2) search. We evaluate end-to-end on Structured3D, with calibrated noise on Gibson, and on real-world author-collected sequences. When sufficient wall geometry is visible, GALoc matches or outperforms depth-based baselines -- achieving 88% sequential localization success at 0.1m over 100-step sequences on Gibson vs the baseline's 68% -- while abstaining in structure-blind scenes.