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 Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Video deraining is an important issue for outdoor vision systems and has beeninvestigated extensively. However, designing optimal architectures by theaggregating model formation and data distribution is a challenging task forvideo deraining. In this paper, we develop a model-guided triple-leveloptimization framework to deduce network architecture with cooperatingoptimization and auto-searching mechanism, named Triple-level Model InferredCooperating Searching (TMICS), for dealing with various video raincircumstances. In particular, to mitigate the problem that existing methodscannot cover various rain streaks distribution, we first design ahyper-parameter optimization model about task variable and hyper-parameter.Based on the proposed optimization model, we design a collaborative structurefor video deraining. This structure includes Dominant Network Architecture(DNA) and Companionate Network Architecture (CNA) that is cooperated byintroducing an Attention-based Averaging Scheme (AAS). To better exploreinter-frame information from videos, we introduce a macroscopic structuresearching scheme that searches from Optical Flow Module (OFM) and TemporalGrouping Module (TGM) to help restore latent frame. In addition, we apply thedifferentiable neural architecture searching from a compact candidate set oftask-specific operations to discover desirable rain streaks removalarchitectures automatically. Extensive experiments on various datasetsdemonstrate that our model shows significant improvements in fidelity andtemporal consistency over the state-of-the-art works. Source code is availableat https://github.com/vis-opt-group/TMICS.

Triple-Level Model Inferred Collaborative Network Architecture for Video Deraining