Deep Generative Models: Deterministic Prediction with an Application in Inverse Rendering

Abstract

Deep generative models are stochastic neural networks capable of learning the distribution of data so as to generate new samples. Conditional Variatonal Autoencoder (CVAE) is a powerful deep generative model aiming at maximizing the lower bound of training data log-likelihood. In this structure, there is appropriate regularizer, which makes it applicable for suitably constraining the solution space in solving ill-posed problems and providing high generalization power. Considering the stochastic prediction characteristic in CVAE, depending on the problem at hand, it is desirable to be able to control the uncertainty in CVAE predictions. Therefore, in this paper we analyze the impact of CVAE's condition on the diversity of solutions given by our designed CVAE in 3D shape inverse rendering as a prediction problem. The experimental results using Modelnet10 and Shapenet datasets and comparison with several recent methods show the appropriate performance of our designed CVAE and verify the hypothesis: "The more informative the conditions in terms of object pose are, the less diverse the CVAE predictions are".