|
We address the problem of segmenting 3D scan data into objects or
object classes. Our segmentation framework is based on a subclass of
Markov Networks, or Markov Random Fields (MRFs), called
Associative
Markov Networks, which support efficient graph-cut inference. The AMN
models incorporate a large set of diverse features and enforce the
preference that adjacent scan points have the same classification
label. We use large margin estimation to discriminatively train the
model from a set of labeled scans; as a result we automatically learn
the relative importance of the features for the segmentation
task. |
