Accountable Textual-Visual Chat Learns to Reject Human Instructions in Image Re-creation.
Zhiwei ZhangYuliang LiuPublished in: Trans. Mach. Learn. Res. (2024)
Keyphrases
- visual attributes
- human visual
- image analysis
- human observers
- image data
- image representation
- image features
- image classification
- single image
- image content
- visual perception
- web images
- image collections
- low level
- multiscale
- image segmentation
- segmentation method
- visual appearance
- visual cues
- image regions
- image pixels
- image matching
- multimedia
- region of interest
- input image
- image retrieval
- spatial information
- visual data
- feature points
- high resolution
- static images
- similarity measure
- visually similar
- visual information
- visual concepts
- high level
- visual input
- image processing