Chinese Image Caption Generation via Visual Attention and Topic Modeling.
Maofu LiuHuijun HuLingjun LiYan YuWeili GuanPublished in: IEEE Trans. Cybern. (2022)
Keyphrases
- visual attention
- topic modeling
- salient regions
- input image
- visual attention model
- image data
- saliency map
- multiscale
- attention mechanism
- image features
- topic models
- image retrieval
- biological vision systems
- image classification
- eye movements
- image content
- vision system
- natural scenes
- eye tracking
- visual features
- low level
- text classification
- image representation
- scale invariant
- text mining
- visual input
- higher level
- computer vision
- video retrieval
- co occurrence
- probabilistic model