TL;DR
DrawMotion is a diffusion-based framework that enables intuitive 3D human motion generation using freehand sketches and text, reducing user effort and increasing control over the output.
Contribution
It introduces a novel multi-condition fusion method and a training-free guidance technique for more accurate and user-aligned motion synthesis from sketches and text.
Findings
Reduces user time by approximately 46.7% in motion generation.
Effectively combines text and sketch conditions for improved control.
Demonstrates superior performance through quantitative and user studies.
Abstract
Text-to-motion generation, which translates textual descriptions into human motions, faces the challenge that users often struggle to precisely convey their intended motions through text alone. To address this issue, this paper introduces DrawMotion, an efficient diffusion-based framework designed for multi-condition scenarios. DrawMotion generates motions based on both a conventional text condition and a novel hand-drawing condition, which provide semantic and spatial control over the generated motions, respectively. Specifically, we tackle the fine-grained motion generation task from three perspectives: 1) freehand drawing condition. To accurately capture users' intended motions without requiring tedious textual input, we develop an algorithm to automatically generate hand-drawn stickman sketches across different dataset formats; 2) multi-condition fusion. We propose a Multi-Condition…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Code & Models
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
