Generative Motion In-betweening by Diffusion over Continuous Implicit Representations
For animation and graphics practitioners, this method enables high-quality motion generation from sparse keyframes, addressing a key bottleneck in existing generative approaches.
The paper tackles motion in-betweening, proposing a latent diffusion model over motion implicit neural representations that improves keyframe accuracy and motion continuity, significantly outperforming prior methods in scenarios with few keyframes.
Recent advances in generative models have yielded impressive progress on motion in-betweening, allowing for more complex, varied, and realistic motion transitions. However, recent methods still exhibit noticeable limitations in preserving keyframe information and ensuring motion continuity. In this paper, we propose a novel pipeline and sampling optimization strategy for latent diffusion models (LDM) based on motion implicit neural representations (INR). By establishing a mapping between INR and sparse spatial or temporal information within latent diffusion, our model can sample the INR parameters from extremely sparse and ambiguous keyframe data and reconstruct plausible and smooth motions from the manifold. Our experiments demonstrate the superior performance of our model, which significantly improves motion generation quality in scenarios with few keyframes while ensuring both keyframe accuracy and diversity of in-between motions.