Flow Plugin Network for conditional generation
This addresses the need for efficient conditional generation in applications like image or 3D modeling, but appears incremental as it adapts existing normalizing flow methods.
The paper tackles the problem of controlling attribute-specific generation in existing generative models without retraining, proposing a Flow Plugin Network that enables conditional sampling, though no concrete performance numbers are provided.
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).