Segmentation of RGB-D indoor scenes by stacking random forests and conditional random fields.
Mikkel ThøgersenSergio EscaleraJordi GonzàlezThomas B. MoeslundPublished in: Pattern Recognit. Lett. (2016)
Keyphrases
- conditional random fields
- indoor scenes
- random forests
- outdoor scenes
- segmentation method
- object segmentation
- single image
- random forest
- decision trees
- probabilistic model
- graphical models
- d scene
- segmentation algorithm
- ensemble methods
- random fields
- logistic regression
- depth images
- information extraction
- machine learning algorithms
- spatial layout
- hidden markov models
- higher order
- pairwise
- generative model
- markov random field
- level set
- image segmentation
- human pose
- base classifiers
- scene understanding
- shape prior
- superpixels
- multi label
- bayesian networks
- computer vision
- similarity measure