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/

Zendy Plus

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

Zendy Tools

Zendy Open

In this work, we address the task of SDR videos to HDRvideos(SDRTV-to-HDRTV). Previous approaches use global feature modulation forSDRTV-to-HDRTV. Feature modulation scales and shifts the features in theoriginal feature space, which has limited mapping capability. In addition, theglobal image mapping cannot restore detail in HDR frames due to the luminancedifferences in different regions of SDR frames. To resolve the appeal, wepropose a two-stage solution. The first stage is a hierarchical Dynamic Contextfeature mapping (HDCFM) model. HDCFM learns the SDR frame to HDR frame mappingfunction via hierarchical feature modulation (HME and HM ) module and a dynamiccontext feature transformation (DCT) module. The HME estimates the featuremodulation vector, HM is capable of hierarchical feature modulation, consistingof global feature modulation in series with local feature modulation, and iscapable of adaptive mapping of local image features. The DCT module constructsa feature transformation module in conjunction with the context, which iscapable of adaptively generating a feature transformation matrix for featuremapping. Compared with simple feature scaling and shifting, the DCT module canmap features into a new feature space and thus has a more excellent featuremapping capability. In the second stage, we introduce a patchdiscriminator-based context generation model PDCG to obtain subjective qualityenhancement of over-exposed regions. PDCG can solve the problem that the modelis challenging to train due to the proportion of overexposed regions of theimage. The proposed method can achieve state-of-the-art objective andsubjective quality results. Specifically, HDCFM achieves a PSNR gain of 0.81 dBat a parameter of about 100K. The number of parameters is 1/14th of theprevious state-of-the-art methods. The test code will be released soon.

SDRTV-to-HDRTV via Hierarchical Dynamic Context Feature Mapping