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

Image-based salient object detection (SOD) has been extensively explored inthe past decades. However, SOD on 360$^\circ$ omnidirectional images is lessstudied owing to the lack of datasets with pixel-level annotations. Toward thisend, this paper proposes a 360$^\circ$ image-based SOD dataset that contains500 high-resolution equirectangular images. We collect the representativeequirectangular images from five mainstream 360$^\circ$ video datasets andmanually annotate all objects and regions over these images with precise maskswith a free-viewpoint way. To the best of our knowledge, it is the first publicavailable dataset for salient object detection on 360$^\circ$ scenes. Byobserving this dataset, we find that distortion from projection, large-scalecomplex scene and small salient objects are the most prominent characteristics.Inspired by these foundings, this paper proposes a baseline model for SOD onequirectangular images. In the proposed approach, we construct adistortion-adaptive module to deal with the distortion caused by theequirectangular projection. In addition, a multi-scale contextual integrationblock is introduced to perceive and distinguish the rich scenes and objects inomnidirectional scenes. The whole network is organized in a progressivelymanner with deep supervision. Experimental results show the proposed baselineapproach outperforms the top-performanced state-of-the-art methods on360$^\circ$ SOD dataset. Moreover, benchmarking results of the proposedbaseline approach and other methods on 360$^\circ$ SOD dataset show theproposed dataset is very challenging, which also validate the usefulness of theproposed dataset and approach to boost the development of SOD on 360$^\circ$omnidirectional scenes.

Distortion-adaptive Salient Object Detection in 360$^\circ$ Omnidirectional Images