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

Although deep learning based methods have achieved great progress inunsupervised video object segmentation, difficult scenarios (e.g., visualsimilarity, occlusions, and appearance changing) are still not well-handled. Toalleviate these issues, we propose a novel Focus on Foreground Network (F2Net),which delves into the intra-inter frame details for the foreground objects andthus effectively improve the segmentation performance. Specifically, ourproposed network consists of three main parts: Siamese Encoder Module, CenterGuiding Appearance Diffusion Module, and Dynamic Information Fusion Module.Firstly, we take a siamese encoder to extract the feature representations ofpaired frames (reference frame and current frame). Then, a Center GuidingAppearance Diffusion Module is designed to capture the inter-frame feature(dense correspondences between reference frame and current frame), intra-framefeature (dense correspondences in current frame), and original semantic featureof current frame. Specifically, we establish a Center Prediction Branch topredict the center location of the foreground object in current frame andleverage the center point information as spatial guidance prior to enhance theinter-frame and intra-frame feature extraction, and thus the featurerepresentation considerably focus on the foreground objects. Finally, wepropose a Dynamic Information Fusion Module to automatically select relativelyimportant features through three aforementioned different level features.Extensive experiments on DAVIS2016, Youtube-object, and FBMS datasets show thatour proposed F2Net achieves the state-of-the-art performance with significantimprovement.

F2Net: Learning to Focus on the Foreground for Unsupervised Video Object Segmentation