BiMotion: B-spline Motion for Text-guided Dynamic 3D Character Generation

TL;DR

BiMotion introduces a novel B-spline based approach for text-guided dynamic 3D character motion generation, enabling high-quality, expressive, and coherent motions that better reflect textual descriptions with faster processing.

Contribution

The paper proposes a continuous B-spline motion representation and a new dataset, improving motion quality and prompt alignment over existing methods without altering the generative model.

Findings

Outperforms state-of-the-art methods in motion quality and prompt adherence

Achieves faster motion generation

Provides a new dataset with diverse, high-quality motion-text pairs

Abstract

Text-guided dynamic 3D character generation has advanced rapidly, yet producing high-quality motion that faithfully reflects rich textual descriptions remains challenging. Existing methods tend to generate limited sub-actions or incoherent motion due to fixed-length temporal inputs and discrete frame-wise representations that fail to capture rich motion semantics. We address these limitations by representing motion with continuous differentiable B-spline curves, enabling more effective motion generation without modifying the capabilities of the underlying generative model. Specifically, our closed-form, Laplacian-regularized B-spline solver efficiently compresses variable-length motion sequences into compact representations with a fixed number of control points. Further, we introduce a normal-fusion strategy for input shape adherence along with correspondence-aware and local-rigidity losses for motion-restoration quality. To train our model, we collate BIMO, a new dataset containing diverse variable-length 3D motion sequences with rich, high-quality text annotations. Extensive evaluations show that our feed-forward framework BiMotion generates more expressive, higher-quality, and better prompt-aligned motions than existing state-of-the-art methods, while also achieving faster generation. Our project page is at: https://wangmiaowei.github.io/BiMotion.github.io/.