ChatPaper.aiChatPaper

可学习的SMPLify:一种无需优化的神经人体姿态逆向运动学解决方案

Learnable SMPLify: A Neural Solution for Optimization-Free Human Pose Inverse Kinematics

August 19, 2025
作者: Yuchen Yang, Linfeng Dong, Wei Wang, Zhihang Zhong, Xiao Sun
cs.AI

摘要

在三维人体姿态与形状估计领域,SMPLify 仍是一个通过迭代优化解决逆向运动学(IK)问题的坚实基准。然而,其高昂的计算成本限制了实际应用。近期跨领域的研究表明,用数据驱动的神经网络替代迭代优化,能在不牺牲精度的前提下显著提升运行效率。受此趋势启发,我们提出了可学习的 SMPLify,这是一个将 SMPLify 中的迭代拟合过程替换为单次回归模型的神经框架。我们的框架设计针对神经 IK 中的两大核心挑战:数据构建与泛化能力。为实现有效训练,我们提出了一种时间采样策略,从连续帧中构建初始化-目标对。为提升对多样化动作及未见姿态的泛化能力,我们采用了以人为中心的归一化方案及残差学习,以缩小解空间。可学习的 SMPLify 既支持序列推理,也可作为插件后处理工具,用于精炼现有的基于图像的估计器。大量实验证明,我们的方法确立了一个实用且简洁的基准:相比 SMPLify,其运行速度提升了近 200 倍,在 3DPW 和 RICH 数据集上展现出良好的泛化性能,且作为插件工具应用于 LucidAction 时,保持了模型无关性。代码已发布于 https://github.com/Charrrrrlie/Learnable-SMPLify。
English
In 3D human pose and shape estimation, SMPLify remains a robust baseline that solves inverse kinematics (IK) through iterative optimization. However, its high computational cost limits its practicality. Recent advances across domains have shown that replacing iterative optimization with data-driven neural networks can achieve significant runtime improvements without sacrificing accuracy. Motivated by this trend, we propose Learnable SMPLify, a neural framework that replaces the iterative fitting process in SMPLify with a single-pass regression model. The design of our framework targets two core challenges in neural IK: data construction and generalization. To enable effective training, we propose a temporal sampling strategy that constructs initialization-target pairs from sequential frames. To improve generalization across diverse motions and unseen poses, we propose a human-centric normalization scheme and residual learning to narrow the solution space. Learnable SMPLify supports both sequential inference and plug-in post-processing to refine existing image-based estimators. Extensive experiments demonstrate that our method establishes itself as a practical and simple baseline: it achieves nearly 200x faster runtime compared to SMPLify, generalizes well to unseen 3DPW and RICH, and operates in a model-agnostic manner when used as a plug-in tool on LucidAction. The code is available at https://github.com/Charrrrrlie/Learnable-SMPLify.
PDF12August 25, 2025