Generating hyper-realistic human images through a unified framework, HyperHuman, utilizing latent structural diffusion for high-quality and diverse results.
Minecraft-ify는 Minecraft 비디오 게임을 위한 캐릭터 텍스처 생성 시스템을 제안하며, StyleGAN 및 StyleCLIP을 활용하여 텍스트로 조작하는 기능을 제공합니다.
本文提出了一種基於區域圖元解構的方法,用於解釋圖像生成神經網絡的內部表徵結構,並證明了每個特徵組件都與特定圖像區域的生成存在明確的對應關係。
ToddlerDiffusion is a novel image generation model that decomposes the generation process into modality-specific stages (sketch, palette, RGB image), enabling efficient training, faster sampling, and interactive editing capabilities, outperforming traditional single-stage diffusion models like LDM.