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

Many video instance segmentation (VIS) methods partition a video sequenceinto individual frames to detect and segment objects frame by frame. However,such a frame-in frame-out (FiFo) pipeline is ineffective to exploit thetemporal information. Based on the fact that adjacent frames in a short clipare highly coherent in content, we propose to extend the one-stage FiFoframework to a clip-in clip-out (CiCo) one, which performs VIS clip by clip.Specifically, we stack FPN features of all frames in a short video clip tobuild a spatio-temporal feature cube, and replace the 2D conv layers in theprediction heads and the mask branch with 3D conv layers, forming clip-levelprediction heads (CPH) and clip-level mask heads (CMH). Then the clip-levelmasks of an instance can be generated by feeding its box-level predictions fromCPH and clip-level features from CMH into a small fully convolutional network.A clip-level segmentation loss is proposed to ensure that the generatedinstance masks are temporally coherent in the clip. The proposed CiCo strategyis free of inter-frame alignment, and can be easily embedded into existing FiFobased VIS approaches. To validate the generality and effectiveness of our CiCostrategy, we apply it to two representative FiFo methods, Yolact\cite{bolya2019yolact} and CondInst \cite{tian2020conditional}, resulting intwo new one-stage VIS models, namely CiCo-Yolact and CiCo-CondInst, whichachieve 37.1/37.3\%, 35.2/35.4\% and 17.2/18.0\% mask AP using the ResNet50backbone, and 41.8/41.4\%, 38.0/38.9\% and 18.0/18.2\% mask AP using the SwinTransformer tiny backbone on YouTube-VIS 2019, 2021 and OVIS valid sets,respectively, recording new state-of-the-arts. Code and video demos of CiCo canbe found at \url{https://github.com/MinghanLi/CiCo}.

One-stage Video Instance Segmentation: From Frame-in Frame-out to Clip-in Clip-out