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

Controllable human image generation (HIG) has numerous real-lifeapplications. State-of-the-art solutions, such as ControlNet and T2I-Adapter,introduce an additional learnable branch on top of the frozen pre-trainedstable diffusion (SD) model, which can enforce various conditions, includingskeleton guidance of HIG. While such a plug-and-play approach is appealing, theinevitable and uncertain conflicts between the original images produced fromthe frozen SD branch and the given condition incur significant challenges forthe learnable branch, which essentially conducts image feature editing forcondition enforcement. In this work, we propose a native skeleton-guideddiffusion model for controllable HIG called HumanSD. Instead of performingimage editing with dual-branch diffusion, we fine-tune the original SD modelusing a novel heatmap-guided denoising loss. This strategy effectively andefficiently strengthens the given skeleton condition during model trainingwhile mitigating the catastrophic forgetting effects. HumanSD is fine-tuned onthe assembly of three large-scale human-centric datasets with text-image-poseinformation, two of which are established in this work. As shown in Figure 1,HumanSD outperforms ControlNet in terms of accurate pose control and imagequality, particularly when the given skeleton guidance is sophisticated.

HumanSD: A Native Skeleton-Guided Diffusion Model for Human Image Generation