Floorplan-Aware Camera Poses Refinement

Processing large indoor scenes is a challenging task, as scan registration and camera trajectory estimation methods accumulate errors across time. As a result, the quality of reconstructed scans is insufficient for some applications, such as visual-based localization and navigation, where the correct position of walls is crucial. For many indoor scenes, there exists an image of a technical ftoorplan that contains information about the geometry and main structural elements of the scene, such as walls, partitions, and doors. We argue that such a ftoorplan is a useful source of spatial information, which can guide a 3D model optimization. The standard RGB-D 3D reconstruction pipeline consists of a tracking module applied to an RGB-D sequence and a bundle adjustment (BA) module that takes the posed RGB-D sequence and corrects the camera poses to improve consistency. We propose a novel optimization algorithm expanding conventional BA that leverages the prior knowledge about the scene structure in the form of a ftoorplan. Our experiments on the Redwood dataset and our self-captured data demonstrate that utilizing ftoorplan improves accuracy of 3D reconstructions.

Authors

• Anna Sokolova
• Filipp Nikitin
• Anna Vorontsova
• Anton Konushin

Keywords

• Bundle adjustment
• Solid modeling
• Navigation
• Robot vision systems
• Pipelines
• Cameras
• Trajectory
• Camera Pose
• 3D Reconstruction
• Reconstruction Accuracy
• Redwood
• Scene Structure
• Cost Function
• Minimum Distance
• General Structure
• Point Cloud
• 3D Space
• Semantic Segmentation
• Depth Map
• 3D Point
• Nearest Point
• Back Projection
• Nearest Neighbor Distance
• Direction Of Gravity
• Reprojection Error
• Geometric Terms
• Plane Wall
• Geometric Consistency
• 3D Error
• Floor Surface
• Scene Model
• Set Of Planes
• Repetitive Patterns