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 introduce UprightNet, a learning-based approach for estimating 2DoF cameraorientation from a single RGB image of an indoor scene. Unlike recent methodsthat leverage deep learning to perform black-box regression from image toorientation parameters, we propose an end-to-end framework that incorporatesexplicit geometric reasoning. In particular, we design a network that predictstwo representations of scene geometry, in both the local camera and globalreference coordinate systems, and solves for the camera orientation as therotation that best aligns these two predictions via a differentiable leastsquares module. This network can be trained end-to-end, and can be supervisedwith both ground truth camera poses and intermediate representations of surfacegeometry. We evaluate UprightNet on the single-image camera orientation task onsynthetic and real datasets, and show significant improvements over priorstate-of-the-art approaches.

UprightNet: Geometry-Aware Camera Orientation Estimation from Single Images