Feb. 28, 2024, 5:46 a.m. | Lvmin Zhang, Maneesh Agrawala

cs.CV updates on arXiv.org arxiv.org

arXiv:2402.17113v1 Announce Type: new
Abstract: We present LayerDiffusion, an approach enabling large-scale pretrained latent diffusion models to generate transparent images. The method allows generation of single transparent images or of multiple transparent layers. The method learns a "latent transparency" that encodes alpha channel transparency into the latent manifold of a pretrained latent diffusion model. It preserves the production-ready quality of the large diffusion model by regulating the added transparency as a latent offset with minimal changes to the original latent …

abstract alpha arxiv cs.cv cs.gr diffusion diffusion models enabling generate image images latent diffusion models layer manifold multiple scale transparency type

Artificial Intelligence – Bioinformatic Expert

@ University of Texas Medical Branch | Galveston, TX

Lead Developer (AI)

@ Cere Network | San Francisco, US

Research Engineer

@ Allora Labs | Remote

Ecosystem Manager

@ Allora Labs | Remote

Founding AI Engineer, Agents

@ Occam AI | New York

AI Engineer Intern, Agents

@ Occam AI | US