Cognitive resilience: Unraveling the proficiency of image-captioning models to interpret masked visual content.
Zhicheng DuZhaotian XieHuazhang YingLikun ZhangPeiwu QinPublished in: Tiny Papers @ ICLR (2024)
Keyphrases
- visual content
- image content
- image collections
- image retrieval
- content based image
- image data
- image features
- visual appearance
- low level
- visual features
- visual descriptors
- textual and visual information
- image representation
- visual concepts
- image classification
- image regions
- video retrieval
- input image
- visual information
- spatial relationships
- image database
- multiscale
- image segmentation
- key frames
- image set
- high resolution
- image annotation
- data processing
- probabilistic model
- high level