Example-based 3D-model synthesis

Shape synthesis is generally a task that is hard to master. New methods using 3D-model synthesis from examples emerged with the growing number of shapes in model repositories. This work reviews and compares three such methods -- one using geometry analysis (Jain et al. 2012), second one reshuffling certain part arrangements (Zheng et al. 2013) and third one employing probabilistic model learned from training datasets Kalogerakis et al. 2012). One of the methods is chosen for implementation, which is the main focus of the work.