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

In this paper, we propose a novel framework named DRL-CPG to learndisentangled latent representation for controllable person image generation,which can produce realistic person images with desired poses and humanattributes (e.g., pose, head, upper clothes, and pants) provided by varioussource persons. Unlike the existing works leveraging the semantic masks toobtain the representation of each component, we propose to generatedisentangled latent code via a novel attribute encoder with transformerstrained in a manner of curriculum learning from a relatively easy step to agradually hard one. A random component mask-agnostic strategy is introduced torandomly remove component masks from the person segmentation masks, which aimsat increasing the difficulty of training and promoting the transformer encoderto recognize the underlying boundaries between each component. This enables themodel to transfer both the shape and texture of the components. Furthermore, wepropose a novel attribute decoder network to integrate multi-level attributes(e.g., the structure feature and the attribute representation) withwell-designed Dual Adaptive Denormalization (DAD) residual blocks. Extensiveexperiments strongly demonstrate that the proposed approach is able to transferboth the texture and shape of different human parts and yield realisticresults. To our knowledge, we are the first to learn disentangled latentrepresentations with transformers for person image generation.

Disentangled Representation Learning for Controllable Person Image Generation