arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2407.02272cs.CVcs.GR
收藏

将人体运动生成与人类感知对齐

Aligning Human Motion Generation with Human Perceptions

Peking University(北京大学);Huawei Technologies, Ltd.(华为技术有限公司)

Haoru Wang, Wentao Zhu, Luyi Miao, Yishu Xu, Feng Gao, Qi Tian, Yizhou Wang

AI总结:

本文提出数据驱动的MotionPercept数据集和MotionCritic评论模型,以人类感知对齐评估与改进人体运动生成质量。

摘要:

人体运动生成是一项具有广泛应用的关键任务。生成运动的高真实感要求自然性、平滑性和合理性。尽管该领域发展迅速,但当前的生成方法往往难以达到这些目标。此外,现有的评估指标通常依赖于基于真值(ground truth)的误差、简单的启发式规则或分布距离,这些指标与人类对运动质量的感知并不一致。在这项工作中,我们提出了一种数据驱动的方法来弥合这一差距,通过引入一个大规模的人类感知评估数据集MotionPercept和一个人类运动评论模型MotionCritic,来捕捉人类感知偏好。我们的评论模型为评估运动质量提供了一个更准确的指标,并且可以方便地集成到运动生成流程中以提高生成质量。大量实验证明了我们的方法在评估和提高生成人体运动质量方面的有效性,通过使其与人类感知对齐。代码和数据公开可用,网址为https://motioncritic.github.io/。

英文摘要:

Human motion generation is a critical task with a wide range of applications. Achieving high realism in generated motions requires naturalness, smoothness, and plausibility. Despite rapid advancements in the field, current generation methods often fall short of these goals. Furthermore, existing evaluation metrics typically rely on ground-truth-based errors, simple heuristics, or distribution distances, which do not align well with human perceptions of motion quality. In this work, we propose a data-driven approach to bridge this gap by introducing a large-scale human perceptual evaluation dataset, MotionPercept, and a human motion critic model, MotionCritic, that capture human perceptual preferences. Our critic model offers a more accurate metric for assessing motion quality and could be readily integrated into the motion generation pipeline to enhance generation quality. Extensive experiments demonstrate the effectiveness of our approach in both evaluating and improving the quality of generated human motions by aligning with human perceptions. Code and data are publicly available at https://motioncritic.github.io/.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑