OVI-MAP:Open-Vocabulary Instance-Semantic Mapping

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Deng, Zilong, Tombari, Federico, Pollefeys, Marc, Wald, Johanna, Barath, Daniel
Natura: Preprint
Pubblicazione: 2026
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866917364543520768
author Deng, Zilong
Tombari, Federico
Pollefeys, Marc
Wald, Johanna
Barath, Daniel
author_facet Deng, Zilong
Tombari, Federico
Pollefeys, Marc
Wald, Johanna
Barath, Daniel
contents Incremental open-vocabulary 3D instance-semantic mapping is essential for autonomous agents operating in complex everyday environments. However, it remains challenging due to the need for robust instance segmentation, real-time processing, and flexible open-set reasoning. Existing methods often rely on the closed-set assumption or dense per-pixel language fusion, which limits scalability and temporal consistency. We introduce OVI-MAP that decouples instance reconstruction from semantic inference. We propose to build a class-agnostic 3D instance map that is incrementally constructed from RGB-D input, while semantic features are extracted only from a small set of automatically selected views using vision-language models. This design enables stable instance tracking and zero-shot semantic labeling throughout online exploration. Our system operates in real time and outperforms state-of-the-art open-vocabulary mapping baselines on standard benchmarks.
format Preprint
id arxiv_https___arxiv_org_abs_2603_26541
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle OVI-MAP:Open-Vocabulary Instance-Semantic Mapping
Deng, Zilong
Tombari, Federico
Pollefeys, Marc
Wald, Johanna
Barath, Daniel
Computer Vision and Pattern Recognition
Incremental open-vocabulary 3D instance-semantic mapping is essential for autonomous agents operating in complex everyday environments. However, it remains challenging due to the need for robust instance segmentation, real-time processing, and flexible open-set reasoning. Existing methods often rely on the closed-set assumption or dense per-pixel language fusion, which limits scalability and temporal consistency. We introduce OVI-MAP that decouples instance reconstruction from semantic inference. We propose to build a class-agnostic 3D instance map that is incrementally constructed from RGB-D input, while semantic features are extracted only from a small set of automatically selected views using vision-language models. This design enables stable instance tracking and zero-shot semantic labeling throughout online exploration. Our system operates in real time and outperforms state-of-the-art open-vocabulary mapping baselines on standard benchmarks.
title OVI-MAP:Open-Vocabulary Instance-Semantic Mapping
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2603.26541