Effective Semantic Pixel Labelling With Convolutional Networks and Conditional Random Fields

Sakrapee Paisitkriangkrai, Jamie Sherrah, Pranam Janney, Anton Van-Den Hengel; Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, 2015, pp. 36-43

Abstract

Large amounts of available training data and increasing computing power have led to the recent success of deep convolutional neural networks (CNN) on a large number of applications. In this paper, we propose an effective semantic pixel labelling using CNN features, hand-crafted features and Conditional Random Fields (CRFs). Both CNN and hand-crafted features are applied to dense image patches to produce per-pixel class probabilities. The CRF infers a labelling that smooths regions while respecting the edges present in the imagery. The method is applied to the ISPRS 2D semantic labelling challenge dataset with competitive classification accuracy.

Related Material

[pdf]
[bibtex]
@InProceedings{Paisitkriangkrai_2015_CVPR_Workshops,
author = {Paisitkriangkrai, Sakrapee and Sherrah, Jamie and Janney, Pranam and Van-Den Hengel, Anton},
title = {Effective Semantic Pixel Labelling With Convolutional Networks and Conditional Random Fields},
booktitle = {Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR) Workshops},
month = {June},
year = {2015}
}