TLControl: Trajectory and Language Control for Human Motion Synthesis

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Wan, Weilin, Dou, Zhiyang, Komura, Taku, Wang, Wenping, Jayaraman, Dinesh, Liu, Lingjie
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909266053431296
author Wan, Weilin
Dou, Zhiyang
Komura, Taku
Wang, Wenping
Jayaraman, Dinesh
Liu, Lingjie
author_facet Wan, Weilin
Dou, Zhiyang
Komura, Taku
Wang, Wenping
Jayaraman, Dinesh
Liu, Lingjie
contents Controllable human motion synthesis is essential for applications in AR/VR, gaming and embodied AI. Existing methods often focus solely on either language or full trajectory control, lacking precision in synthesizing motions aligned with user-specified trajectories, especially for multi-joint control. To address these issues, we present TLControl, a novel method for realistic human motion synthesis, incorporating both low-level Trajectory and high-level Language semantics controls, through the integration of neural-based and optimization-based techniques. Specifically, we begin with training a VQ-VAE for a compact and well-structured latent motion space organized by body parts. We then propose a Masked Trajectories Transformer (MTT) for predicting a motion distribution conditioned on language and trajectory. Once trained, we use MTT to sample initial motion predictions given user-specified partial trajectories and text descriptions as conditioning. Finally, we introduce a test-time optimization to refine these coarse predictions for precise trajectory control, which offers flexibility by allowing users to specify various optimization goals and ensures high runtime efficiency. Comprehensive experiments show that TLControl significantly outperforms the state-of-the-art in trajectory accuracy and time efficiency, making it practical for interactive and high-quality animation generation.
format Preprint
id arxiv_https___arxiv_org_abs_2311_17135
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle TLControl: Trajectory and Language Control for Human Motion Synthesis
Wan, Weilin
Dou, Zhiyang
Komura, Taku
Wang, Wenping
Jayaraman, Dinesh
Liu, Lingjie
Computer Vision and Pattern Recognition
Graphics
Controllable human motion synthesis is essential for applications in AR/VR, gaming and embodied AI. Existing methods often focus solely on either language or full trajectory control, lacking precision in synthesizing motions aligned with user-specified trajectories, especially for multi-joint control. To address these issues, we present TLControl, a novel method for realistic human motion synthesis, incorporating both low-level Trajectory and high-level Language semantics controls, through the integration of neural-based and optimization-based techniques. Specifically, we begin with training a VQ-VAE for a compact and well-structured latent motion space organized by body parts. We then propose a Masked Trajectories Transformer (MTT) for predicting a motion distribution conditioned on language and trajectory. Once trained, we use MTT to sample initial motion predictions given user-specified partial trajectories and text descriptions as conditioning. Finally, we introduce a test-time optimization to refine these coarse predictions for precise trajectory control, which offers flexibility by allowing users to specify various optimization goals and ensures high runtime efficiency. Comprehensive experiments show that TLControl significantly outperforms the state-of-the-art in trajectory accuracy and time efficiency, making it practical for interactive and high-quality animation generation.
title TLControl: Trajectory and Language Control for Human Motion Synthesis
topic Computer Vision and Pattern Recognition
Graphics
url https://arxiv.org/abs/2311.17135