Zero-shot spatial layout conditioning for text-to-image diffusion models.
Guillaume CouaironMarlène CareilMatthieu CordStéphane LathuilièreJakob VerbeekPublished in: CoRR (2023)
Keyphrases
- spatial layout
- scene recognition
- input image
- multiscale
- image data
- image classification
- image content
- single image
- image representation
- diffusion models
- indoor environments
- image features
- low level
- bag of words
- scene classification
- high resolution
- object classes
- feature points
- edge detection
- image regions
- text mining
- visual words
- keywords