Towards Practical Single-shot Motion Synthesis

TL;DR

This paper presents a faster, more efficient GAN-based method for single-shot motion synthesis that achieves competitive quality and diversity, with the ability to generate and compose motion in a single forward pass.

Contribution

We introduce an improved GAN architecture for single-shot motion generation that reduces training time and enhances diversity, enabling real-time motion synthesis and composition.

Findings

Achieves up to 6.8x faster training than baseline GANs.

Maintains competitive quality and diversity on the Mixamo benchmark.

Enables motion mixing and composition with a single forward pass.

Abstract

Despite the recent advances in the so-called "cold start" generation from text prompts, their needs in data and computing resources, as well as the ambiguities around intellectual property and privacy concerns pose certain counterarguments for their utility. An interesting and relatively unexplored alternative has been the introduction of unconditional synthesis from a single sample, which has led to interesting generative applications. In this paper we focus on single-shot motion generation and more specifically on accelerating the training time of a Generative Adversarial Network (GAN). In particular, we tackle the challenge of GAN's equilibrium collapse when using mini-batch training by carefully annealing the weights of the loss functions that prevent mode collapse. Additionally, we perform statistical analysis in the generator and discriminator models to identify correlations between training stages and enable transfer learning. Our improved GAN achieves competitive quality and diversity on the Mixamo benchmark when compared to the original GAN architecture and a single-shot diffusion model, while being up to x6.8 faster in training time from the former and x1.75 from the latter. Finally, we demonstrate the ability of our improved GAN to mix and compose motion with a single forward pass. Project page available at https://moverseai.github.io/single-shot.