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

360{\deg} images are usually represented in either equirectangular projection(ERP) or multiple perspective projections. Different from the flat 2D images,the detection task is challenging for 360{\deg} images due to the distortion ofERP and the inefficiency of perspective projections. However, existing methodsmostly focus on one of the above representations instead of both, leading tolimited detection performance. Moreover, the lack of appropriate bounding-boxannotations as well as the annotated datasets further increases thedifficulties of the detection task. In this paper, we present a standard objectdetection framework for 360{\deg} images. Specifically, we adapt theterminologies of the traditional object detection task to the omnidirectionalscenarios, and propose a novel two-stage object detector, i.e., ReprojectionR-CNN by combining both ERP and perspective projection. Owing to theomnidirectional field-of-view of ERP, Reprojection R-CNN first generates coarseregion proposals efficiently by a distortion-aware spherical region proposalnetwork. Then, it leverages the distortion-free perspective projection andrefines the proposed regions by a novel reprojection network. We construct twonovel synthetic datasets for training and evaluation. Experiments reveal thatReprojection R-CNN outperforms the previous state-of-the-art methods on the mAPmetric. In addition, the proposed detector could run at 178ms per image in thepanoramic datasets, which implies its practicability in real-worldapplications.

Reprojection R-CNN: A Fast and Accurate Object Detector for 360{\deg} Images