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

Unsupervised video object segmentation (VOS) aims to detect the most salientobject in a video sequence at the pixel level. In unsupervised VOS, moststate-of-the-art methods leverage motion cues obtained from optical flow mapsin addition to appearance cues to exploit the property that salient objectsusually have distinctive movements compared to the background. However, as theyare overly dependent on motion cues, which may be unreliable in some cases,they cannot achieve stable prediction. To reduce this motion dependency ofexisting two-stream VOS methods, we propose a novel motion-as-option networkthat optionally utilizes motion cues. Additionally, to fully exploit theproperty of the proposed network that motion is not always required, weintroduce a collaborative network learning strategy. On all the publicbenchmark datasets, our proposed network affords state-of-the-art performancewith real-time inference speed.

Treating Motion as Option to Reduce Motion Dependency in Unsupervised Video Object Segmentation