Conditioned Masked Language and Image Modeling for Image-Text Dense Retrieval.
Ziyang LuoYadong XiRongsheng ZhangGongZheng LiZeng ZhaoJing MaPublished in: EMNLP (Findings) (2022)
Keyphrases
- image data
- single image
- image retrieval
- input image
- image features
- template matching
- image representation
- low level
- edge detection
- image analysis
- multiscale
- image set
- image matching
- image content
- high resolution
- web images
- image description
- image classification
- image segmentation
- image collections
- region of interest
- image pixels
- information retrieval
- scanned documents
- segmentation method
- segmentation algorithm
- text mining
- similarity measure
- image sequences
- image processing