Model based probabilistic piecewise curve approximation

In this work, we approach the piecewise curve approximation problem with a model-based probabilistic framework. For this purpose, we propose three different models. These models can be used for feature extraction or compression. The first model is a variant of the Bayesian regression model where we can parametrically alter the design matrix. The second model approaches the piecewise curve approximation as a clustering problem. The third model adds temporal connectivity to the second model and combines Hidden Markov models with linear regression. We run the first and the third models on a curve which is used to rank existing algorithms and show that our approaches outperforms its rivals. We also run our models on several real-life curves to show their capabilities.