From Uncertainty to Determinism: Coarse-to-Fine Visual Floorplan Localization without Ray Matching

2026-07-29

AI Summary

This work addresses the challenge of multimodal pose distribution in visual floorplan localization, which arises from cross-modal information asymmetry and repetitive indoor layouts. To this end, we propose a coarse-to-fine localization framework that eliminates the need for ray matching. Our approach introduces, for the first time, an image-conditioned pose diffusion model to capture continuous multimodal pose distributions, complemented by a local refinement module that achieves sub-meter accuracy within candidate regions. By employing candidate-centered floorplan cropping and enabling end-to-end inference without offline preprocessing, the system attains state-of-the-art accuracy and robustness on the S3D (full) and ZInD benchmarks, departing from conventional paradigms that rely on precomputed maps or lookup tables during testing.

Abstract

Visual Floorplan Localization (FLoc) has emerged as a promising solution for indoor localization by matching egocentric images against minimalist structural maps. However, due to cross-modal information asymmetry and repetitive indoor layouts, visual FLoc is fundamentally challenged by multimodal pose distributions, where visually identical observations map to distinct, spatially separated locations. Existing ray-matching-based methods tackle this by explicitly predicting sparse geometric or semantic rays, which inherently incur information loss and demand resource-intensive preprocessing alongside exhaustive matching during inference. In this paper, we bypass the intermediate ray-matching paradigm and propose a coarse-to-fine visual FLoc framework that progresses from uncertainty to determinism. In the coarse stage, we design an image-conditioned pose diffusion model to parameterize the continuous multimodal pose distribution, effectively routing stochastically initialized pose particles toward distinct candidate modes. In the refinement stage, we propose a localized refiner that predicts bounded sub-meter pose residuals from candidate-centered floorplan crops, where structural ambiguities are largely eliminated. Our method effectively balances global multi-hypothesis tracking and local sub-meter refinement without requiring any offline map preprocessing or test-time lookup tables. Comprehensive results on the S3D (full) and ZInD benchmarks demonstrate that our approach achieves state-of-the-art accuracy and robustness.

Problem

• Visual Floorplan Localization
• multimodal pose distribution
• indoor localization
• cross-modal asymmetry
• structural ambiguity

Innovation

• pose diffusion
• coarse-to-fine localization
• visual floorplan localization
• multimodal pose distribution
• ray-free matching