Abstract 3D shape generation is a challenging problem due to the high-dimensional output space and complex part confifigurations of real-world objects. As a result, existing algorithms experience diffificulties in accurate generative modeling of 3D shapes. Here, we propose a novel factorized generative model for 3D shape generation that sequentially transitions from coarse to fifine scale shape generation. To this end, we introduce an unsupervised primitive discovery algorithm based on a higher-order conditional random fifield model. Using the primitive parts for shapes as attributes, a parameterized 3D representation is modeled in the fifirst stage. This representation is further refifined in the next stage by adding fifine scale details to shape. Our results demonstrate improved representation ability of the generative model and better quality samples of newly generated 3D shapes. Further, our primitive generation approach can accurately parse common objects into a simplifified representation