MaskGAN is a novel framework for diverse and interactive facial image manipulation. It leverages semantic masks as an intermediate representation to enable flexible and high-fidelity face manipulation. The key components of MaskGAN are the Dense Mapping Network (DMN) and Editing Behavior Simulated Training (EBST). DMN learns the style mapping between a user-modified mask and a target image, allowing for diverse generation results. EBST models the user's editing behavior on the source mask, enhancing the robustness of the framework to various manipulated inputs. The framework is evaluated on two challenging tasks: attribute transfer and style copy, demonstrating superior performance compared to state-of-the-art methods. To facilitate large-scale studies, a high-resolution face dataset with fine-grained mask annotations, named CelebAMask-HQ, is constructed. The dataset includes over 30,000 face images with 19 facial component categories. MaskGAN provides an intuitive interface for users to interactively edit facial features, shape, and accessories, making it a powerful tool for facial image manipulation.MaskGAN is a novel framework for diverse and interactive facial image manipulation. It leverages semantic masks as an intermediate representation to enable flexible and high-fidelity face manipulation. The key components of MaskGAN are the Dense Mapping Network (DMN) and Editing Behavior Simulated Training (EBST). DMN learns the style mapping between a user-modified mask and a target image, allowing for diverse generation results. EBST models the user's editing behavior on the source mask, enhancing the robustness of the framework to various manipulated inputs. The framework is evaluated on two challenging tasks: attribute transfer and style copy, demonstrating superior performance compared to state-of-the-art methods. To facilitate large-scale studies, a high-resolution face dataset with fine-grained mask annotations, named CelebAMask-HQ, is constructed. The dataset includes over 30,000 face images with 19 facial component categories. MaskGAN provides an intuitive interface for users to interactively edit facial features, shape, and accessories, making it a powerful tool for facial image manipulation.