From Open Waters to Enclosed Cabins: ProteusVPR for Cross-Scene Visual Place Recognition in Maritime Perception and Cabin Inspection

A new visual place recognition framework called ProteusVPR promises to improve autonomous robotic inspection aboard ships, where stark shifts between open decks and enclosed cabins have long confounded existing navigation systems [1]. The two-stage retrieval-refinement model was detailed in a paper submitted to arXiv on June 23, 2026 [1]. Autonomous robotic inspection in maritime environments presents unique challenges for Visual Place Recognition (VPR) because robots must transition between visually distinct domains: open decks with sparse textures and severe illumination changes, and enclosed cabins with repetitive structures and high visual ambiguity [1]. Existing VPR methods, designed primarily for urban or indoor scenes, fail to generalize reliably across these starkly different scenarios [1]. To address this, the researchers propose ProteusVPR [1]. The first stage employs any standard VPR model for initial image retrieval. The second stage introduces a geometric-visual estimation network that fuses the retrieved image with two temporally preceding frames, incorporating geometric descriptors, a local affine coordinate system, and camera azimuth encoding to achieve precise localization [1]. To support the task, the team introduced the XHZ dataset, an 8K-panoramic ship-borne dataset collected from an operational vessel, featuring multi-floor cabin structures, deck transition zones, and strict query-database separation for rigorous evaluation [1]. Extensive experiments on the XHZ dataset demonstrate that ProteusVPR consistently improves the localization accuracy across multiple VPR backbones, reducing mean localization error by over 60% on average [1]. The paper was posted on arXiv, an open-access repository of electronic preprints that, as of November 2024, receives about 24,000 submissions per month and is not peer reviewed [6]. The abstract page for the paper includes experimental tools developed through arXivLabs, a framework that allows collaborators to develop and share new arXiv features directly on the website [4]. arXivLabs was launched as a formalized framework in 2020 to enable community innovation while ensuring partners share arXiv’s values of openness, community, excellence, and user data privacy [4]. The Labs tabs on the article page include tools such as the Bibliographic Explorer, which displays citation information, and the CORE Recommender, which facilitates exploration of relevant open access papers from a global network of research repositories [5].