HDNet: Human-like discrimination with visual key for few-shot cross-domain object detection.
Mao Zhen LiuXiaoguang DiWenzhuang WangPublished in: Knowl. Based Syst. (2024)
Keyphrases
- cross domain
- object detection
- visual features
- transfer learning
- multiple domains
- domain adaptation
- text categorization
- knowledge transfer
- sentiment classification
- visual information
- cross domain learning
- computer vision
- video data
- video content
- video sequences
- feature selection
- web image annotation
- knowledge discovery
- object categories
- object recognition
- target domain
- face recognition
- neural network