CatVersion: Concatenating Embeddings for Diffusion-Based Text-to-Image Personalization.
Ruoyu ZhaoMingrui ZhuShiyin DongNannan WangXinbo GaoPublished in: CoRR (2023)
Keyphrases
- image data
- single image
- input image
- image features
- image content
- multiscale
- template matching
- image representation
- information retrieval
- low level
- image segmentation
- image retrieval
- scanned documents
- high resolution
- edge detection
- diffusion process
- web images
- similarity measure
- image pixels
- segmentation method
- image regions
- image collections
- edge map
- pixel values
- text graphics
- test images
- hough transform
- segmentation algorithm
- feature points
- image classification
- dimensionality reduction
- image processing
- search engine