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

Video instance segmentation (VIS) is a new and critical task in computervision. To date, top-performing VIS methods extend the two-stage Mask R-CNN byadding a tracking branch, leaving plenty of room for improvement. In contrast,we approach the VIS task from a new perspective and propose a one-stage spatialgranularity network (SG-Net). Compared to the conventional two-stage methods,SG-Net demonstrates four advantages: 1) Our method has a one-stage compactarchitecture and each task head (detection, segmentation, and tracking) iscrafted interdependently so they can effectively share features and enjoy thejoint optimization; 2) Our mask prediction is dynamically performed on thesub-regions of each detected instance, leading to high-quality masks of finegranularity; 3) Each of our task predictions avoids using expensiveproposal-based RoI features, resulting in much reduced runtime complexity perinstance; 4) Our tracking head models objects centerness movements fortracking, which effectively enhances the tracking robustness to differentobject appearances. In evaluation, we present state-of-the-art comparisons onthe YouTube-VIS dataset. Extensive experiments demonstrate that our compactone-stage method can achieve improved performance in both accuracy andinference speed. We hope our SG-Net could serve as a strong and flexiblebaseline for the VIS task. Our code will be available.

SG-Net: Spatial Granularity Network for One-Stage Video Instance Segmentation