An Unsupervised Conformal Learning Algorithm For Image Segmentation

Details An Unsupervised Conformal Learning Algorithm For Image Segmentation An Unsupervised Conformal Learning Algorithm For Image Segmentation

2011-01-19

arxiv.org/abs/1101.3755v1

Computer Science

Computer Vision and Pattern Recognition

14 pages, 4 figures

Scientific paper

Conformal learners have been resent developments in the area of learning in a random world. Until now, for them to work in online prediction mode, supervised labels need to be present with the objects or examples that are under investigation. The manuscript proposes a simple unsupervised conformal learning algorithm that works by coupling the creation of new classes and agglomeration of examples into these classes, on the basis of old already seen examples. The online generated clusters are probabilistically bounded via multivariate Chebyshev inequality and are stable in a finite data set, with a converging error rate and the number of clusters. The domain of static image segmentation is used to demostrate the efficacy of the conformal prediction in the case of unsupervised learning.

Affiliated with

Also associated with

No associations

Rating

An Unsupervised Conformal Learning Algorithm For Image Segmentation does not yet have a rating. At this time, there are no reviews or comments for this scientific paper.

If you have personal experience with An Unsupervised Conformal Learning Algorithm For Image Segmentation, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and An Unsupervised Conformal Learning Algorithm For Image Segmentation will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFWR-SCP-O-157357