Generative models have gained many researchers' attention in the last years resulting in models such as StyleGAN for human face generation or PointFlow for the 3D point cloud generation. However, by default, we cannot control its sampling process, i.e., we cannot generate a sample with a specific set of attributes. The current approach is model retraining with additional inputs and different architecture, which requires time and computational resources. We propose a novel approach that enables to a generation of objects with a given set of attributes without retraining the base model. For this purpose, we utilize the normalizing flow models - Conditional Masked Autoregressive Flow and Conditional Real NVP, as a Flow Plugin Network (FPN).
翻译:在过去的几年里,生成模型引起了许多研究人员的注意,从而产生了3D点云生成的StyleGAN等模型或PointFlow等3D点云生成模型。然而,在默认情况下,我们无法控制其取样过程,即我们不能用一套特定属性生成样本。目前的方法是用额外的投入和不同的结构进行模型再培训,这需要时间和计算资源。我们提出了一个新颖的方法,使具有一套特定属性的物体能够生成,而无需对基准模型进行再培训。为此,我们使用正常化的流程模型 - 有条件的蒙面自动侵蚀流动和条件性真实的NVP 网络(FPN 网络 ) 。