Harnessing the Spatial-Temporal Attention of Diffusion Models for High-Fidelity Text-to-Image Synthesis.
Qiucheng WuYujian LiuHandong ZhaoTrung BuiZhe LinYang ZhangShiyu ChangPublished in: ICCV (2023)
Keyphrases
- image synthesis
- spatial temporal
- high fidelity
- diffusion models
- computer graphics
- real time
- diffusion model
- temporal information
- spatio temporal
- high quality
- action recognition
- spatial and temporal
- high resolution
- information retrieval
- social networks
- video shots
- keywords
- machine learning
- text mining
- text documents
- spatial information
- information diffusion
- human actions
- computer vision
- semantic information
- image data