Graph adjacency matrix learning for irregularly sampled Markovian natural images.
Stefania ColonneseMauro BiagiRoberto CusaniGaetano ScaranoPublished in: EUSIPCO (2017)
Keyphrases
- natural images
- adjacency matrix
- multiscale
- graph representation
- higher order
- object recognition
- graph matching
- image denoising
- denoising
- unsupervised learning
- input image
- directed graph
- graph clustering
- bipartite graph
- image patches
- statistics of natural images
- undirected graph
- machine learning
- superpixels
- weighted graph
- graph structure
- sparse coding
- pairwise
- learning algorithm