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

Sparse depth measurements are widely available in many applications such asaugmented reality, visual inertial odometry and robots equipped with low costdepth sensors. Although such sparse depth samples work well for certainapplications like motion tracking, a complete depth map is usually preferredfor broader applications, such as 3D object recognition, 3D reconstruction andautonomous driving. Despite the recent advancements in depth prediction fromsingle RGB images with deeper neural networks, the existing approaches do notyield reliable results for practical use. In this work, we propose a neuralnetwork with post-optimization, which takes an RGB image and sparse depthsamples as input and predicts the complete depth map. We make three majorcontributions to advance the state-of-the-art: an improved backbone networkarchitecture named EDNet, a semantic edge-weighted loss function and a semanticmesh deformation optimization method. Our evaluation results outperform theexisting work consistently on both indoor and outdoor datasets, and itsignificantly reduces the mean average error by up to 19.5% under the samesettings of 200 sparse samples on NYU-Depth-V2 dataset.

Sparse Depth Completion with Semantic Mesh Deformation Optimization