Keyword-driven image captioning via Context-dependent Bilateral LSTM.
Xiaodan ZhangShengfeng HeXinhang SongPengxu WeiShuqiang JiangQixiang YeJianbin JiaoRynson W. H. LauPublished in: ICME (2017)
Keyphrases
- context dependent
- low level
- input image
- image features
- semantic level
- image data
- single image
- multiscale
- image analysis
- image representation
- image content
- image classification
- test images
- image segmentation
- high resolution
- region of interest
- image pixels
- context free
- edge detection
- segmentation method
- medical images
- template matching
- image structure
- image retrieval
- keywords
- feature points
- segmentation algorithm
- image regions
- pixel values