Parameter and Data Efficient Spectral Style-DCGAN

Aryan Garg

Keywords: Generative modeling, unconditional generation, data efficient, parameter efficient, machine learning, animal face generation

TL;DR: A simple, highly parameter, and data-efficient adversarial network for unconditional animal face generation

Abstract: We present a simple, highly parameter, and data-efficient adversarial network for unconditional face generation. Our method: Spectral Style-DCGAN or SSD utilizes only 6.574 million parameters and 4739 dog faces from the Animal Faces HQ (AFHQ) dataset as training samples while preserving fidelity at low resolutions up to 64x64. Code available at Anonymous-repo.

Submission Number: 22