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

We propose a novel dense mapping framework for sparse visual SLAM systemswhich leverages a compact scene representation. State-of-the-art sparse visualSLAM systems provide accurate and reliable estimates of the camera trajectoryand locations of landmarks. While these sparse maps are useful forlocalization, they cannot be used for other tasks such as obstacle avoidance orscene understanding. In this paper we propose a dense mapping framework tocomplement sparse visual SLAM systems which takes as input the camera poses,keyframes and sparse points produced by the SLAM system and predicts a densedepth image for every keyframe. We build on CodeSLAM and use a variationalautoencoder (VAE) which is conditioned on intensity, sparse depth andreprojection error images from sparse SLAM to predict an uncertainty-awaredense depth map. The use of a VAE then enables us to refine the dense depthimages through multi-view optimization which improves the consistency ofoverlapping frames. Our mapper runs in a separate thread in parallel to theSLAM system in a loosely coupled manner. This flexible design allows forintegration with arbitrary metric sparse SLAM systems without delaying the mainSLAM process. Our dense mapper can be used not only for local mapping but alsoglobally consistent dense 3D reconstruction through TSDF fusion. We demonstrateour system running with ORB-SLAM3 and show accurate dense depth estimationwhich could enable applications such as robotics and augmented reality.

CodeMapping: Real-Time Dense Mapping for Sparse SLAM using Compact Scene Representations