English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Understanding how humans leverage semantic knowledge to navigate unfamiliarenvironments and decide where to explore next is pivotal for developing robotscapable of human-like search behaviors. We introduce a zero-shot navigationapproach, Vision-Language Frontier Maps (VLFM), which is inspired by humanreasoning and designed to navigate towards unseen semantic objects in novelenvironments. VLFM builds occupancy maps from depth observations to identifyfrontiers, and leverages RGB observations and a pre-trained vision-languagemodel to generate a language-grounded value map. VLFM then uses this map toidentify the most promising frontier to explore for finding an instance of agiven target object category. We evaluate VLFM in photo-realistic environmentsfrom the Gibson, Habitat-Matterport 3D (HM3D), and Matterport 3D (MP3D)datasets within the Habitat simulator. Remarkably, VLFM achievesstate-of-the-art results on all three datasets as measured by success weightedby path length (SPL) for the Object Goal Navigation task. Furthermore, we showthat VLFM's zero-shot nature enables it to be readily deployed on real-worldrobots such as the Boston Dynamics Spot mobile manipulation platform. We deployVLFM on Spot and demonstrate its capability to efficiently navigate to targetobjects within an office building in the real world, without any priorknowledge of the environment. The accomplishments of VLFM underscore thepromising potential of vision-language models in advancing the field ofsemantic navigation. Videos of real-world deployment can be viewed atnaoki.io/vlfm.

VLFM: Vision-Language Frontier Maps for Zero-Shot Semantic Navigation