Passage-Aware Structural Mapping for RGB-D Visual SLAM

2026-04-27

AI Summary

This work addresses the limitation of existing RGB-D visual SLAM systems in effectively modeling traversable structures such as doorways, which hinders semantic navigation and integration with Building Information Modeling (BIM). To bridge this gap, the paper introduces, for the first time, a structured representation of doorways within visual SLAM. By fusing geometric discontinuities, semantic segmentation, plane fitting, and traversal-based evidence from camera-wall interactions, the method infers doorway locations and traversability, abstracting them as passage nodes embedded into an enhanced scene graph. Built upon the vS-Graphs framework, the approach demonstrates robust doorway detection and connectivity modeling in indoor office environments, laying the groundwork for semantic alignment between SLAM and BIM. The implementation is publicly released.

Technology Category

• Computer Vision: Visual Reasoning & Symbolic Representations
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Web query analysis, representation and understanding

Abstract

Doorways and passages are critical structural elements for indoor robot navigation, yet they remain underexplored in modern Visual SLAM (VSLAM) frameworks. This paper presents a passage-aware structural mapping approach for RGB-D VSLAM that detects doors and traversable openings by jointly fusing geometric, semantic, and topological cues. Doors are modeled as planar entities embedded within walls and classified as traversable or non-traversable based on their coplanarity with the supporting wall. Passages are inferred through two complementary strategies: traversal evidence accumulated from camera-wall interactions across consecutive keyframes, and geometric opening validation based on discontinuities in the mapped wall geometry. The proposed method is integrated into vS-Graphs as a proof of concept, enriching its scene graph with passage-level abstractions and improving room connectivity modeling. Qualitative evaluations on indoor office sequences demonstrate reliable doorway detection, and the framework lays the foundation for exploiting these elements in BIM-informed VSLAM. The source code is publicly available at https://github.com/snt-arg/visual_sgraphs/tree/doorway_integration.

Problem

• Visual SLAM
• door detection
• passage modeling
• indoor navigation
• structural mapping

Innovation

• passage-aware mapping
• RGB-D SLAM
• door detection
• structural abstraction
• scene graph