1、 IEEE Industry Connections (IEEEIC) Landmarks and Measurement Standards Comparison in 3D Bodymodel Processing IEEE-SA Industry Connections White Paper Authors: Carol McDonald, Gneiss Concept Yingying Wu, Kansas State University Alfredo Ballester, Instituto de Biomecnica, Universistat Politcnica de V
2、alncia Michael Stahl, Intel Core Contributors: John Fijen (Tc2 Labs) Dongsoo Han (Zelus) Julianne Harris (Target) Sean Inyong Jeon (CLO Virtual Fashion, Inc.) Makiko Kouchi (National Institute of AIST Japan) Amit Kumar (Gerber) Luciano Oviedo (University of Warwick) Emma Scott (Fashion Should Empowe
3、r) Daniel Shor (Browzwear) IEEE | 3 Park Avenue | New York, NY 10016-5997 | USA 1 Copyright 2018 IEEE. All rights reserved. IEEE Industry Connections (IEEEIC) Landmarks and Measurement Standards Comparison in 3D Bodymodel Processing Authors: Carol McDonald, Gneiss Concept Yingying Wu, Kansas State U
4、niversity Alfredo Ballester, Instituto de Biomecnica, Universistat Politcnica de Valncia Michael Stahl, Intel Core Contributors: John Fijen (Tc2 Labs) Dongsoo Han (Zelus) Julianne Harris (Target) Sean Inyong Jeon (CLO Virtual Fashion, Inc.) Makiko Kouchi (National Institute of AIST Japan) Amit Kumar
5、 (Gerber) Luciano Oviedo (University of Warwick) Emma Scott (Fashion Should Empower) Daniel Shor (Browzwear) 2 Copyright 2018 IEEE. All rights reserved. Trademarks and Disclaimers IEEE believes the information in this publication is accurate as of its publication date; such information is subject to
6、 change without notice. IEEE is not responsible for any inadvertent errors. The Institute of Electrical and Electronics Engineers, Inc. 3 Park Avenue, New York, NY 100165997, USA Copyright 2018 by The Institute of Electrical and Electronics Engineers, Inc. All rights reserved. Published February 201
7、8. Printed in the United States of America. IEEE is a registered trademark in the U. S. Patent fitness for a particular purpose; noninfringement; quality, accuracy, effectiveness, currency, or completeness of the Work or content within the Work. In addition, IEEE and the author and contributors disc
8、laim any and all conditions relating to: results; and workmanlike effort. This document is supplied “AS IS” and “WITH ALL FAULTS.” Although the author and contributors who have created this Work believe that the information and guidance given in this Work serve as an enhancement to users, all person
9、s must rely upon their own skill and judgment when making use of it. IN NO EVENT SHALL IEEE OR THE AUTHOR AND CONTRIBUTORS BE LIABLE FOR ANY ERRORS OR OMISSIONS OR DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO: PROCUREMENT OF SUBSTITUTE GOO
10、DS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS WORK, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE
11、 AND REGARDLESS OF WHETHER SUCH DAMAGE WAS FORESEEABLE. Further, information contained in this Work may be protected by intellectual property rights held by third parties or organizations, and the use of this information may require the user to negotiate with any such rights holders in order to lega
12、lly acquire the rights to do so. IEEE and the author and contributors make no assurances that the use of the material contained in this work is free from patent infringement. Essential Patent Claims may exist for which no assurances have been made to the IEEE, whether by participants in this white p
13、aper activity or entities outside the activity. The IEEE is not responsible for identifying essential patent claims for which a license may be required, for conducting inquiries into the legal validity or scope of patents claims, or determining whether any licensing terms or conditions, if any, or a
14、ny licensing agreements are reasonable or nondiscriminatory. Users are expressly advised that determination of the validity of any patent rights, and the risk of infringement of such rights, is entirely their own responsibility. No commitment to grant licenses under patent rights on a reasonable or
15、nondiscriminatory basis has been sought or received from any rights holder. The policies and procedures under which this document was created can be viewed at http:/standards.ieee.org/about/sasb/iccom/. This Work is published with the understanding that IEEE and the author and contributors are suppl
16、ying information through this Work, not attempting to render engineering or other professional services. If such services are required, the assistance of an appropriate professional should be sought. IEEE is not responsible for the statements and opinions advanced in this Work. 4 Copyright 2018 IEEE
17、. All rights reserved. Contents ABSTRACT . 5 1. Introduction . 5 2. The role of landmarks and measurements in 3DBP applications 7 CAD clothing design . 8 CAD footwear design and fitting 8 3. Landscape of existing standards 9 4. Detailed comparison of preferred landmarks and measurements from use cas
18、es 11 Required landmarks from use cases for full body . 12 Body regions 15 Required digital landmarks from use cases for full body 16 Digital body regions . 19 5. Summary and recommendations 20 6. Citations . 22 Appendix A Summary of content of 3DBPrelated standards 24 ISO 72501 (2017) 24 ISO 85591
19、(2017) 24 ISO 20685:2010 24 ISO 206852:2015 . 25 International Society for the Advancement of Kinanthropometry (ISAK) . 25 IWare Laboratory 25 ISO/TS 19408:2015 25 Appendix B Landmarks and measurements from the existing standards 26 Landmarks from all standards . 26 Measurements from all standards .
20、 28 5 Copyright 2018 IEEE. All rights reserved. IEEE Industry Connections (IEEEIC) Landmarks and Measurement Standards Comparison in 3D Bodymodel Processing Abstract This white paper reviews the current standards landscape for threedimensional body scanning, body landmarking, and measuring. Internat
21、ional standards are compared and recommendations are made for a minimal set of landmarks and measurements (L ISO 72501:2017 8), for these applications, definitions are still based on manually palpated landmarks and do not consider the particularities and advantages of the digital methods. This means
22、 that each measuring software developer creates a different interpretation and implementation of the same measurements and landmarks. To address this issue, new standards have been recently released providing body measurements, landmarks, body part and joint definitions from a natively digital persp
23、ective for digital fashion applications (ISO 188251:2016 10, ISO 188252:2016 11). Despite the fact that these standards constitute an important step forward, these definitions are conceived to work with digital human models rather than with actual body scans of people. Additionally, none of the prev
24、iously mentioned standards considers a procedure for defining new measurements, which is common practice among body scanner and measuring software suppliers. Regarding quality, in the case of manual measurements, the skill of measurer strongly affects the reliability of manual measurements (Kouchi,
25、et al. 17); moreover, compatibility among manual measurements is often compromised due to systematic errors between differently skilled measurers. Analogously, in the case of digitally elicited measurements, the quality of the 3D scans has a strong influence on the reliability of measurements; moreo
26、ver, the compatibility among digital measurements is often compromised due to differences in the interpretation and implementations of different measuring software. The main benefit of 3D scanners is that they provide the whole surface of the body, i.e., 3D body model, which cannot be provided by tr
27、aditional anthropometry. The accuracy of the raw scan point cloud (or surface) acquired by a body scan of a rigid object is also addressed by ISO 206852:2015 12 and by proprietary procedures developed by manufacturers. However, it is not resolved for measuring living humans, which are soft and artic
28、ulated. Moreover, the surface manipulations along the processing pipeline from the raw scan to the 3D body model is neither tracked nor quantified. Within 3DBP initiative, the quality subgroup intends to provide methods, tools, benchmarks, resources, and testing procedures to define and quantify the
29、 quality of 3D models, as well as the quality of the critical metadata for use cases, such as body landmarks and measurements. Quality quantification is intended to be part of the qualityrelated metadata, which will provide complementary information about what the user receives and to what extent it
30、 is reliable, accurate, and trustworthy. Among the different steps within the 3D body processing pipeline (Figure 1), the subgroup is initially focused on (A) 3D scanning, (B) mesh surface reconstruction, 1International Standards Organization 2Internatoinal Society for the Advancement in Kinanthropo
31、metry 3International Electrotechnical Comission 4Commit Europen de Normalisation 5American Society for the Advancement in Kinanthropometry 7 Copyright 2018 IEEE. All rights reserved. (C) digital landmarking, and (D) digital measuring since they are the more relevant for the considered use cases. Fig
32、ure 1 : 3D Body Data Processing Pipeline (McDonald 20) Within these four processes, the quality attributes considered so far are related to the following: Descriptive information about the process, e.g., scanner supplier, scanner specifications, software version, scanning pose, or scanning attire. Q
33、ualitative descriptors and quantitative metrics for the processed 3D surfaces related to noise, artifacts, redundancies, holes, smoothing, and surface reconstruction. Reliability of body measuring software. Compatibility of digital body measuring and landmarking to methods dependent upon different d
34、igital software and/or traditional methods. This white paper provides some examples of the role of measurements in 3DBP use cases as well as discusses the current landscape of standards related to body measurements and landmarks, and identifies the gaps not covered by them. Moreover, it provides a s
35、hortlist of measurements and landmarks that will be used within the initial use cases considered by 3DBP initiative; i.e., fit and size estimation, retail, clothing manufacturing, CAD tool developers, and body model storage and service. 2. The role of landmarks and measurements in 3DBP applications
36、As noted in the first paper published by the 3DBP Industry Connections group 24, there are different building blocks that can make up a 3DBP solution, such as the following: a. Scanning or acquisition of the 3D surface data b. 3D Model generation c. Landmarks and measurements (L some applications do
37、 require accurate landmark placement to provide good draping results. CAD footwear design and fitting Another example for the use of L or nonweight bearing, NWB). ISO/TS 19408 15 states that measurements should be taken in an evenly weighted position between the two feet (half weight bearing, HWB).
38、However, that may not be the best position for all use cases in footwear, especially if the underside of the foot should not be compressed. The foot measurements maximum allowable error between extracted value and traditionally measured value should be within 2 mm (ISO 20685:2010 12). This is in the
39、 same order of magnitude as the differences between the different weightbearing conditions for different foot measurements, which can range from 37 mm depending on the measurement (Houston, et al. 6, Oladipo 21,Telfer and Woodburn 22, Tsung, et al. 23, Xiong, et al. 25). Additional critical factors
40、affecting L other landmarks share the same names but have varied levels of differences in their definitions. Required landmarks from use cases for full body The standards compared have the L in such cases, it is recommended to use the largest or highest side and record which side is utilized. PLEASE
41、 NOTE: There are known discrepancies between standard definitions. Where direct comparisons between these standards could not be found, definitions with a close correlation were chosen. Table 2 : List of landmarks and reference to the standard describing the definition Terminolgy6Recommended standar
42、d ISO 85591 ISO 72501 ISAK 1 Armpit Back Fold Point (L sitting) Defines the maximum mean difference allowed between the scanner results and the anthropometrist results, to be considered sufficiently accurate. Defines the sample size of scanned people to meet a 95% confidence interval for the mean di
43、fference between measured and scanned measurements. Lists recommended clothing to use while scanning. Provides a protocol for evaluating the comparability between manual and scanderived measurements. ISO 206852:2015 Ergonomics3D scanning methodologies for internationally compatible anthropometric da
44、tabasesPart 2: Evaluation protocol of surface shape and repeatability of relative landmark positions This part of ISO 20685 addresses protocols for testing of 3D surfacescanning systems in the acquisition of human body shape data and measurements. The standard applies to the landmark positions deter
45、mined by an anthropometrist. It does not apply to landmark positions automatically calculated by software from the point cloud. The tests are aimed at verifying: The intrinsic accuracy of the scanner using a knownsize ball placed at various locations in the scan volume The repeatability of landmark
46、placement when the subject stands at slightly different positions within the scan volume International Society for the Advancement of Kinanthropometry (ISAK) International Standards for Anthropometric Assessment (a guide) ISAK created a comprehensive anthropometric guide. They offer anthropometrist
47、training classes and certifications at a number of levels. IWare Laboratory Four pages of diagrams for foot landmarks and measurements ISO/TS 19408:2015 FootwearSizingVocabulary and terminology This technical specification defines terms commonly used for measuring feet and lasts and for determining
48、the size of footwear. Refers to ISO 19407 FootwearSizingConversion of major sizing systems and ISO 19952, FootwearVocabulary Section 2.1 defines Foot dimensions and shoe sizing and Section 2.2 defines Last dimensions 26 Copyright 2018 IEEE. All rights reserved. Appendix B Landmarks and measurements
49、from the existing standards Landmarks from all standards ISO 72501: 2017 5.2 Acromion 5.3 Cervicale 5.4 Crotch level 5.5 Ectocanthus 5.6 Glabella 5.7 Illiospinale anterius 5.8 Lowest point on rib cage 5.9 Menton 5.10 Mesosternale 5.11 Nuchale 5.12 Olecranon 5.13 Orbitale 5.14 Opisthocranion 5.15 Sellion 5.16 Stylion (radial) 5.17 Suprapatella, sitting 5.18 Thelion 5.19 Tibiale 5.20 Tragion 5.21 Ulnar Stylion 5.22 Vertex (top of head) ISO 85591:2017 3.1.1 Shoulder point 3.1.2 Cen
