Publication Details
Overview
 
 
 

Chapter in Book/ Report/ Conference proceeding

Abstract 

Accurate hand measurement data is of crucial importance in medicalscience, fashion industry, and augmented/virtual reality applications. Conventional methods extract the hand measurements manually using a measuring tape, thereby being very time-consumingand yielding unreliable measurements. In this paper, we propose–tothe best of our knowledge–the first deep-learning-based methodto automatically measure the hand in a non-contact manner froma single 3D hand scan. The proposed method employs a 3D handscan, extracts the features, reconstructs the hand by making useof a 3D hand template, transfers the measurements defined on thetemplate and extracts them from the reconstructed hand. In orderto train, validate, and test the method, a novel large-scale synthetichand dataset is generated. The results on both the unseen syntheticdata and the unseen real scans captured by the Occipital structuresensor Mark I demonstrate that the proposed method outperformsthe state-of-the-art method in most hand measurement types

Reference