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

Flow based garment warping is an integral part of image-based virtual try-onnetworks. However, optimizing a single flow predicting network for simultaneousglobal boundary alignment and local texture preservation results in sub-optimalflow fields. Moreover, dense flows are inherently not suited to handleintricate conditions like garment occlusion by body parts or by other garments.Forcing flows to handle the above issues results in various distortions liketexture squeezing, and stretching. In this work, we propose a novel approachwhere we disentangle the global boundary alignment and local texture preservingtasks via our GlobalNet and LocalNet modules. A consistency loss is thenemployed between the two modules which harmonizes the local flows with theglobal boundary alignment. Additionally, we explicitly handle occlusions bypredicting body-parts visibility mask, which is used to mask out the occludedregions in the warped garment. The masking prevents the LocalNet frompredicting flows that distort texture to compensate for occlusions. We alsointroduce a novel regularization loss (NIPR), that defines a criteria toidentify the regions in the warped garment where texture integrity is violated(squeezed or stretched). NIPR subsequently penalizes the flow in those regionsto ensure regular and coherent warps that preserve the texture in localneighborhoods. Evaluation on a widely used virtual try-on dataset demonstratesstrong performance of our network compared to the current SOTA methods.

GC-VTON: Predicting Globally Consistent and Occlusion Aware Local Flows with Neighborhood Integrity Preservation for Virtual Try-on