Generating Pixel Art Character Sprites using GANs
This addresses the effort-intensive process for game designers, but it is incremental as it builds on existing Pix2Pix methods.
The paper tackled the problem of generating pixel art character sprite sheets for game development by using conditional generative adversarial networks, achieving promising results with models that sometimes produced images very close to ground truth as measured by FID.
Iterating on creating pixel art character sprite sheets is essential to the game development process. However, it can take a lot of effort until the final versions containing different poses and animation clips are achieved. This paper investigates using conditional generative adversarial networks to aid the designers in creating such sprite sheets. We propose an architecture based on Pix2Pix to generate images of characters facing a target side (e.g., right) given sprites of them in a source pose (e.g., front). Experiments with small pixel art datasets yielded promising results, resulting in models with varying degrees of generalization, sometimes capable of generating images very close to the ground truth. We analyze the results through visual inspection and quantitatively with FID.