Injecting Semantic Concepts into End-to-End Image Captioning.
Zhiyuan FangJianfeng WangXiaowei HuLin LiangZhe GanLijuan WangYezhou YangZicheng LiuPublished in: CoRR (2021)
Keyphrases
- end to end
- semantic concepts
- image collections
- image data
- image content
- semantic gap
- image regions
- visual concepts
- image analysis
- low level
- input image
- image representation
- image features
- multiscale
- image retrieval
- low level features
- visual information
- image segmentation
- image classification
- multi modal
- image annotation
- congestion control
- semantic concept detection
- spatial information
- spatial relations
- natural language processing
- video analysis
- object recognition
- high level
- metadata