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

Video instance segmentation (VIS) aims at segmenting and tracking objects invideos. Prior methods typically generate frame-level or clip-level objectinstances first and then associate them by either additional tracking heads orcomplex instance matching algorithms. This explicit instance associationapproach increases system complexity and fails to fully exploit temporal cuesin videos. In this paper, we design a simple, fast and yet effectivequery-based framework for online VIS. Relying on an instance query and proposalpropagation mechanism with several specially developed components, thisframework can perform accurate instance association implicitly. Specifically,we generate frame-level object instances based on a set of instancequery-proposal pairs propagated from previous frames. This instancequery-proposal pair is learned to bind with one specific object across framesthrough conscientiously developed strategies. When using such a pair to predictan object instance on the current frame, not only the generated instance isautomatically associated with its precursors on previous frames, but the modelgets a good prior for predicting the same object. In this way, we naturallyachieve implicit instance association in parallel with segmentation andelegantly take advantage of temporal clues in videos. To show the effectivenessof our method InsPro, we evaluate it on two popular VIS benchmarks, i.e.,YouTube-VIS 2019 and YouTube-VIS 2021. Without bells-and-whistles, our InsProwith ResNet-50 backbone achieves 43.2 AP and 37.6 AP on these two benchmarksrespectively, outperforming all other online VIS methods.

InsPro: Propagating Instance Query and Proposal for Online Video Instance Segmentation