PulseAugur
LIVE 13:04:07
research · [3 sources] ·
0
research

COMPASS algorithm uses floor plans for robot visual localization

By PulseAugur Editorial · Summary by gemini-2.5-flash-lite from 3 sources

Researchers have developed COMPASS, a novel algorithm for robot localization that leverages both geometric and semantic information from architectural floor plans. This method utilizes a multi-channel radial descriptor to encode environmental layouts, drawing inspiration from LiDAR-based place recognition techniques. COMPASS aims to improve pose estimation for robots equipped with dual fisheye cameras by integrating visual data with floor plan priors, demonstrating feasibility through a proof-of-concept matching wall-window patterns. AI

Summary written by gemini-2.5-flash-lite from 3 sources. How we write summaries →

IMPACT Enhances robot navigation accuracy by integrating semantic floor plan data with visual input.

RANK_REASON Academic paper detailing a new algorithm for visual localization.

Read on arXiv cs.CV →

COVERAGE [3]

  1. Hugging Face Daily Papers TIER_1 ·

    COMPASS: COmpact Multi-channel Prior-map And Scene Signature for Floor-Plan-Based Visual Localization

    Architectural floor plans are widely available priors which contain not only geometry but also the semantic information of the environment, yet existing localization methods largely ignore this semantic information. To address this, we present COMPASS, an algorithm that exploits …

  2. arXiv cs.CV TIER_1 · Muhammad Shaheer, Miguel Fernandez-Cortizas, Asier Bikandi-Noya, Holger Voos, Jose Luis Sanchez-Lopez ·

    COMPASS: COmpact Multi-channel Prior-map And Scene Signature for Floor-Plan-Based Visual Localization

    arXiv:2604.25388v1 Announce Type: new Abstract: Architectural floor plans are widely available priors which contain not only geometry but also the semantic information of the environment, yet existing localization methods largely ignore this semantic information. To address this,…

  3. arXiv cs.CV TIER_1 · Jose Luis Sanchez-Lopez ·

    COMPASS: COmpact Multi-channel Prior-map And Scene Signature for Floor-Plan-Based Visual Localization

    Architectural floor plans are widely available priors which contain not only geometry but also the semantic information of the environment, yet existing localization methods largely ignore this semantic information. To address this, we present COMPASS, an algorithm that exploits …

RELATED TOPICS