Effect of Gender, Pose and Camera Distance on Human Body Dimensions Estimation

TL;DR

This study systematically evaluates how gender, pose, and camera distance affect CNN-based human body dimension estimation from images, revealing key factors influencing accuracy and providing insights for improving HBDE methods.

Contribution

It offers a comprehensive assessment of CNN performance in HBDE under various controlled conditions, filling a gap in systematic experimental analysis.

Findings

CNN can successfully estimate human body dimensions from images.

Camera distance significantly impacts estimation accuracy.

Gender and pose influence the error rates in body dimension estimation.

Abstract

Human Body Dimensions Estimation (HBDE) is a task that an intelligent agent can perform to attempt to determine human body information from images (2D) or point clouds or meshes (3D). More specifically, if we define the HBDE problem as inferring human body measurements from images, then HBDE is a difficult, inverse, multi-task regression problem that can be tackled with machine learning techniques, particularly convolutional neural networks (CNN). Despite the community's tremendous effort to advance human shape analysis, there is a lack of systematic experiments to assess CNNs estimation of human body dimensions from images. Our contribution lies in assessing a CNN estimation performance in a series of controlled experiments. To that end, we augment our recently published neural anthropometer dataset by rendering images with different camera distance. We evaluate the network inference absolute and relative mean error between the estimated and actual HBDs. We train and evaluate the CNN in four scenarios: (1) training with subjects of a specific gender, (2) in a specific pose, (3) sparse camera distance and (4) dense camera distance. Not only our experiments demonstrate that the network can perform the task successfully, but also reveal a number of relevant facts that contribute to better understand the task of HBDE.