Real-time multi-view facial landmark detector learned by the structured output SVM.
Michal UricárVojtech FrancDiego ThomasAkihiro SugimotoVáclav HlavácPublished in: FG (2015)
Keyphrases
- multi view
- structured output
- maximum margin
- multiple views
- support vector machine
- latent variables
- depth map
- d objects
- three dimensional
- conditional random fields
- semi supervised
- support vector
- multi task learning
- multi class
- semi supervised learning
- feature vectors
- pairwise
- object recognition
- similarity measure
- computer vision