A unified convolutional neural network integrated with conditional random field for pipe defect segmentation.
Mingzhu WangJack C. P. ChengPublished in: Comput. Aided Civ. Infrastructure Eng. (2020)
Keyphrases
- conditional random fields
- convolutional neural network
- crf model
- image labeling
- semantic segmentation
- segmentation method
- long range
- markov random field
- graphical models
- image segmentation
- segmentation algorithm
- probabilistic model
- higher order
- maximum entropy
- information extraction
- hidden markov models
- level set
- structured prediction
- generative model
- random fields
- multiscale
- face detection
- sequence classification
- undirected graphical models
- graph cuts
- supervised training
- approximate inference
- contextual features
- information retrieval
- pairwise
- superpixels
- object segmentation
- shape prior
- neural network
- medical images
- semi supervised