BriefGPT.xyz
Mar, 2023
姿态引导扩散模型的一致视图合成
Consistent View Synthesis with Pose-Guided Diffusion Models
HTML
PDF
Hung-Yu Tseng, Qinbo Li, Changil Kim, Suhib Alsisan, Jia-Bin Huang...
TL;DR
本文提出了一种姿态引导扩散模型来从单张图像中生成一致的、持久的新视角视频，并设计了一个利用极线作为约束的注意力层以促进不同视角之间的关联。实验结果表明，该模型在合成和真实数据集上表现优异，胜过基于Transformer和GAN的现有方法。
Abstract
novel view synthesis
from a single image has been a cornerstone problem for many
virtual reality applications
that provide immersive experiences. However, most existing techniques can only synthesize novel views
→