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

In this letter, we present a neural field-based real-time monocular mappingframework for accurate and dense Simultaneous Localization and Mapping (SLAM).Recent neural mapping frameworks show promising results, but rely on RGB-D orpose inputs, or cannot run in real-time. To address these limitations, ourapproach integrates dense-SLAM with neural implicit fields. Specifically, ourdense SLAM approach runs parallel tracking and global optimization, while aneural field-based map is constructed incrementally based on the latest SLAMestimates. For the efficient construction of neural fields, we employmulti-resolution grid encoding and signed distance function (SDF)representation. This allows us to keep the map always up-to-date and adaptinstantly to global updates via loop closing. For global consistency, wepropose an efficient Sim(3)-based pose graph bundle adjustment (PGBA) approachto run online loop closing and mitigate the pose and scale drift. To enhancedepth accuracy further, we incorporate learned monocular depth priors. Wepropose a novel joint depth and scale adjustment (JDSA) module to solve thescale ambiguity inherent in depth priors. Extensive evaluations acrosssynthetic and real-world datasets validate that our approach outperformsexisting methods in accuracy and map completeness while preserving real-timeperformance.

HI-SLAM: Monocular Real-time Dense Mapping with Hybrid Implicit Fields