We have witnessed rapid progress on 3D-aware image synthesis, leveraging\nrecent advances in generative visual models and neural rendering. Existing\napproaches however fall short in two ways: first, they may lack an underlying\n3D representation or rely on view-inconsistent rendering, hence synthesizing\nimages that are not multi-view consistent; second, they often depend upon\nrepresentation network architectures that are not expressive enough, and their\nresults thus lack in image quality. We propose a novel generative model, named\nPeriodic Implicit Generative Adversarial Networks ($\\pi$-GAN or pi-GAN), for\nhigh-quality 3D-aware image synthesis. $\\pi$-GAN leverages neural\nrepresentations with periodic activation functions and volumetric rendering to\nrepresent scenes as view-consistent 3D representations with fine detail. The\nproposed approach obtains state-of-the-art results for 3D-aware image synthesis\nwith multiple real and synthetic datasets.\n