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

Diffusion models have achieved remarkable image generation quality surpassingprevious generative models. However, a notable limitation of diffusion models,in comparison to GANs, is their difficulty in smoothly interpolating betweentwo image samples, due to their highly unstructured latent space. Such a smoothinterpolation is intriguing as it naturally serves as a solution for the imagemorphing task with many applications. In this work, we present DiffMorpher, thefirst approach enabling smooth and natural image interpolation using diffusionmodels. Our key idea is to capture the semantics of the two images by fittingtwo LoRAs to them respectively, and interpolate between both the LoRAparameters and the latent noises to ensure a smooth semantic transition, wherecorrespondence automatically emerges without the need for annotation. Inaddition, we propose an attention interpolation and injection technique and anew sampling schedule to further enhance the smoothness between consecutiveimages. Extensive experiments demonstrate that DiffMorpher achieves starklybetter image morphing effects than previous methods across a variety of objectcategories, bridging a critical functional gap that distinguished diffusionmodels from GANs.

DiffMorpher: Unleashing the Capability of Diffusion Models for Image Morphing