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

Dense 3D reconstruction and ego-motion estimation are key challenges inautonomous driving and robotics. Compared to the complex, multi-modal systemsdeployed today, multi-camera systems provide a simpler, low-cost alternative.However, camera-based 3D reconstruction of complex dynamic scenes has provenextremely difficult, as existing solutions often produce incomplete orincoherent results. We propose R3D3, a multi-camera system for dense 3Dreconstruction and ego-motion estimation. Our approach iterates betweengeometric estimation that exploits spatial-temporal information from multiplecameras, and monocular depth refinement. We integrate multi-camera featurecorrelation and dense bundle adjustment operators that yield robust geometricdepth and pose estimates. To improve reconstruction where geometric depth isunreliable, e.g. for moving objects or low-textured regions, we introducelearnable scene priors via a depth refinement network. We show that this designenables a dense, consistent 3D reconstruction of challenging, dynamic outdoorenvironments. Consequently, we achieve state-of-the-art dense depth predictionon the DDAD and NuScenes benchmarks.

R3D3: Dense 3D Reconstruction of Dynamic Scenes from Multiple Cameras