Multi-level Visual Fusion Networks for Image Captioning.
Dongming ZhouCanlong ZhangZhixin LiZhiwen WangPublished in: IJCNN (2020)
Keyphrases
- low level
- image features
- fusion method
- image segmentation
- image data
- input image
- template matching
- visual appearance
- image classification
- image representation
- image regions
- visual perception
- image content
- single image
- visual cues
- image analysis
- image matching
- medical image retrieval
- fused image
- visually similar
- edge detection
- multiscale
- image retrieval
- visual data
- visual features
- human observers
- fusion methods
- human vision
- web images
- image pixels
- test images
- visual information
- data fusion
- pixel values
- object recognition
- image fusion
- network structure
- similarity measure
- human visual
- feature points
- visual attributes
- social networks