Multi-class image segmentation using conditional random fields and global classification

A key aspect of semantic image segmentation is to integrate local and global features for the prediction of local segment labels. We present an approach to multi-class segmentation which combines two methods for this integration: a Conditional Random Field (CRF) which couples to local image features and an image classification method which considers global features. The CRF follows the approach of Reynolds & Murphy (2007) and is based on an unsupervised multi scale pre-segmentation of the image into patches, where patch labels correspond to the random variables of the CRF. The output of the classifier is used to constraint this CRF. We demonstrate and compare the approach on a standard semantic segmentation data set.

• Type: Conference
• Year: 2009
• Where: ICML
• Authors: Nils Plath, Marc Toussaint, Shinichi Nakajima
• Keywords: ICML 2009, Local Image Features, Machine Learning, Semantic Image Segmentation, Standard Semantic Segmentation