Graph Neural Network based Scene Change Detection Using Scene Graph Embedding with Hybrid Classification Loss.
Soyeon KimKyung-No JooChan-Hyun YounPublished in: ICTC (2021)
Keyphrases
- graph embedding
- scene change detection
- feature representation
- dimensionality reduction
- dictionary learning
- classification accuracy
- data representation
- label information
- d scene
- low dimensional
- neural network
- support vector
- semi supervised
- text classification
- support vector machine svm
- pattern recognition
- video sequences
- graph structure
- supervised learning
- feature selection
- feature space
- subspace learning
- discriminant analysis
- machine learning
- moving objects
- decision trees
- image sequences
- visual information
- training samples
- visual features
- unsupervised learning
- training set
- support vector machine