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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

We address the problem of semi-supervised video object segmentation (VOS),where the masks of objects of interests are given in the first frame of aninput video. To deal with challenging cases where objects are occluded ormissing, previous work relies on greedy data association strategies that makedecisions for each frame individually. In this paper, we propose a novelapproach to defer the decision making for a target object in each frame, untila global view can be established with the entire video being taken intoconsideration. Our approach is in the same spirit as Multiple HypothesesTracking (MHT) methods, making several critical adaptations for the VOSproblem. We employ the bounding box (bbox) hypothesis for tracking treeformation, and the multiple hypotheses are spawned by propagating the precedingbbox into the detected bbox proposals within a gated region starting from theinitial object mask in the first frame. The gated region is determined by agating scheme which takes into account a more comprehensive motion model ratherthan the simple Kalman filtering model in traditional MHT. To further designmore customized algorithms tailored for VOS, we develop a novel maskpropagation score instead of the appearance similarity score that could bebrittle due to large deformations. The mask propagation score, together withthe motion score, determines the affinity between the hypotheses during treepruning. Finally, a novel mask merging strategy is employed to handle maskconflicts between objects. Extensive experiments on challenging datasetsdemonstrate the effectiveness of the proposed method, especially in the case ofobject missing.

MHP-VOS: Multiple Hypotheses Propagation for Video Object Segmentation