Improving the quality of K-NN graphs for image databases through vector sparsification

Keyphrases

• Automatic Image Annotation 50%
• Class Label 50%
• Construction Method 50%
• Content-based Image Retrieval 50%
• Efficient Construction 50%
• Feature Vector 50%
• Global Mean 50%
• Graph Construction 50%
• Graph-based 50%
• Image Database 100%
• Individual Image 50%
• K-nearest 100%
• Laplacian Score 50%
• Local Variant 50%
• Neighbor Graph 100%
• Neighbor Set 50%
• Neighborhood Graph 50%
• Noisy Features 50%
• Object Features 50%
• Query Image 50%
• Relevant Images 50%
• Semantic Quality 100%
• Semantically Related 50%
• Sparsification 50%
• Vector Sparsification 100%

Computer Science

• Content-based image retrieval 50%
• Feature Vector 50%
• Graph Construction 50%
• Image Annotation 50%
• Image Database 100%
• Laplacian Score 50%
• Nearest Neighbor Graph 100%
• Neighbourhood Graph 50%