Semantic-visual Guided Transformer for Few-shot Class-incremental Learning.
Wenhao QiuSichao FuJingyi ZhangChengxiang LeiQinmu PengPublished in: CoRR (2023)
Keyphrases
- incremental learning
- visual features
- incremental learning algorithm
- fuzzy artmap
- learning process
- batch mode
- low level
- high level
- sports video
- semantic content
- visual information
- batch learning
- unlabeled samples
- news video
- fuzzy logic
- semi supervised
- neural network
- visual concepts
- semantic similarity
- low level features
- video content
- video shots
- semantic concepts
- video data
- supervised learning
- active learning
- feature vectors
- neural learning
- text classification