Latent Diffusion-GAN: Adversarial Learning in the Autoencoded Latent Space

Simple item page
dc.contributor.authorJun, U-Chae
dc.contributor.authorKo, Jaeeun
dc.contributor.authorKang, Jiwoo
dc.contributor.editorMasia, Belen
dc.contributor.editorThies, Justus
dc.date.accessioned2026-04-17T14:05:50Z
dc.date.available2026-04-17T14:05:50Z
dc.date.issued2026
dc.description.abstractDiffusion models are powerful generative frameworks for producing high-quality images by denoising latent variables from random noise. However, training with likelihood-based objectives can lead to oversmoothed high-frequency details such as textures and sharp edges. Adversarial training with GANs enhances sharpness but usually requires additional discriminator networks. We propose Latent Diffusion Generative Adversarial Networks (LD-GAN), a framework that integrates adversarial learning into diffusion models without modifying their pipeline. LD-GAN leverages the pretrained variational autoencoder as an energy-based discriminator, enabling adversarial training without extra parameters while preserving the latent priors learned from large datasets. We also introduce a structural consistency energy aligning encoder and decoder representations, improving perceptual quality. Experiments show improved sample fidelity, sharpness, and diversity across multiple generation tasks while maintaining efficient training dynamics.
dc.description.number2
dc.description.sectionheadersDiffusion and Beyond: Controlled Image Generation and Stylization
dc.description.seriesinformationComputer Graphics Forum
dc.description.volume45
dc.identifier.doi10.1111/cgf.70409
dc.identifier.issn1467-8659
dc.identifier.pages21 pages
dc.identifier.urihttps://diglib.eg.org/handle/10.1111/cgf70409
dc.identifier.urihttps://doi.org/10.1111/cgf.70409
dc.publisherThe Eurographics Association and John Wiley & Sons Ltd.
dc.rightsCC-BY-4.0
dc.rights.urihttps://creativecommons.org/licenses/by/4.0/
dc.subjectComputer vision
dc.titleLatent Diffusion-GAN: Adversarial Learning in the Autoencoded Latent Space
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
cgf70409.pdf
Size:
5.52 MB
Format:
Adobe Portable Document Format
Download
Collections
45-Issue 2
EG 2026 - Full Papers - CGF 45-Issue 2