DiffDis: Empowering Generative Diffusion Model with Cross-Modal Discrimination Capability.
Runhui HuangJianhua HanGuansong LuXiaodan LiangYihan ZengWei ZhangHang XuPublished in: ICCV (2023)
Keyphrases
- cross modal
- diffusion model
- multi modal
- anisotropic diffusion
- information diffusion
- diffusion process
- generative model
- multimedia retrieval
- multimedia databases
- visual recognition
- image retrieval
- visual data
- diffusion tensor
- image processing
- content based retrieval
- visual similarity
- image analysis
- similarity measure