Divcon: Learning Concept Sequences for Semantically Diverse Image Captioning.
Yue ZhengYa-Li LiShengjin WangPublished in: ICASSP (2023)
Keyphrases
- input image
- image data
- multiscale
- learning process
- image analysis
- image content
- single image
- concept learning
- image retrieval
- high resolution
- image features
- image pixels
- learning tasks
- learning systems
- feature points
- learning algorithm
- edge detection
- low level
- online learning
- active learning
- reinforcement learning
- visual concepts
- image segmentation
- segmentation method
- test images
- pixel values
- inter related
- image classification
- template matching
- image regions
- segmentation algorithm
- supervised learning
- face recognition
- computer vision