Abstract. Recently, there has been increasing interests in applying aspect models (e.g., PLSA and LDA) in image segmentation. However, these models ignore spatial relationships amo...
In this paper we propose a machine learning approach to classify melanocytic lesions in malignant and benign from dermatoscopic images. The image database is composed of 433 benign...
This paper proposes a data driven image segmentation algorithm, based on decomposing the target output (ground truth). Classical pixel labeling methods utilize machine learning al...
Manual reconstruction of ripped-up documents can be a very difficult and time-consuming task. This paper discusses a semi-automatic toolset that can be used for reconstructing rip...
In this work, we propose to improve the neighboring relationship ability of the Hidden Markov Chain (HMC) model, by extending the memory lengthes of both the Markov chain process ...