GCN-Based Multi-Modal Multi-Label Attribute Classification in Anime Illustration Using Domain-Specific Semantic Features.
Ziwen LanKeisuke MaedaTakahiro OgawaMiki HaseyamaPublished in: ICIP (2022)
Keyphrases
- multi modal
- multi label
- semantic features
- text classification
- multi label classification
- image annotation
- image classification
- class labels
- text categorization
- automatic image annotation
- visual features
- classification algorithm
- classification accuracy
- graph cuts
- machine learning
- semantic concepts
- high dimensional
- feature selection
- decision trees
- support vector
- supervised learning
- neural network
- semantic similarity
- feature space
- wordnet
- prior knowledge
- unlabeled data
- image representation
- unsupervised learning
- low level features
- learning algorithm
- feature set
- text mining
- binary classifiers
- data mining
- support vector machine