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

Moving Object Segmentation is a challenging task for jittery/wobbly videos.For jittery videos, the non-smooth camera motion makes discrimination betweenforeground objects and background layers hard to solve. While most recent worksfor moving video object segmentation fail in this scenario, our methodgenerates an accurate segmentation of a single moving object. The proposedmethod performs a sparse segmentation, where frame-wise labels are assignedonly to trajectory coordinates, followed by the pixel-wise labeling of frames.The sparse segmentation involving stabilization and clustering of trajectoriesin a 3-stage iterative process. At the 1st stage, the trajectories areclustered using pairwise Procrustes distance as a cue for creating an affinitymatrix. The 2nd stage performs a block-wise Procrustes analysis of thetrajectories and estimates Frechet means (in Kendall's shape space) of theclusters. The Frechet means represent the average trajectories of the motionclusters. An optimization function has been formulated to stabilize the Frechetmeans, yielding stabilized trajectories at the 3rd stage. The accuracy of themotion clusters are iteratively refined, producing distinct groups ofstabilized trajectories. Next, the labels obtained from the sparse segmentationare propagated for pixel-wise labeling of the frames, using a GraphCut basedenergy formulation. Use of Procrustes analysis and energy minimization inKendall's shape space for moving object segmentation in jittery videos, is thenovelty of this work. Second contribution comes from experiments performed on adataset formed of 20 real-world natural jittery videos, with manually annotatedground truth. Experiments are done with controlled levels of artificial jitteron videos of SegTrack2 dataset. Qualitative and quantitative results indicatethe superiority of the proposed method.

Moving Object Segmentation in Jittery Videos by Stabilizing Trajectories Modeled in Kendall's Shape Space