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

Wide field-of-view (FoV) cameras efficiently capture large portions of thescene, which makes them attractive in multiple domains, such as automotive androbotics. For such applications, estimating depth from multiple images is acritical task, and therefore, a large amount of ground truth (GT) data isavailable. Unfortunately, most of the GT data is for pinhole cameras, making itimpossible to properly train depth estimation models for large-FoV cameras. Wepropose the first method to train a stereo depth estimation model on the widelyavailable pinhole data, and to generalize it to data captured with larger FoVs.Our intuition is simple: We warp the training data to a canonical, large-FoVrepresentation and augment it to allow a single network to reason about diversetypes of distortions that otherwise would prevent generalization. We showstrong generalization ability of our approach on both indoor and outdoordatasets, which was not possible with previous methods.

FoVA-Depth: Field-of-View Agnostic Depth Estimation for Cross-Dataset Generalization