{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,30]],"date-time":"2026-03-30T03:15:19Z","timestamp":1774840519445,"version":"3.50.1"},"reference-count":51,"publisher":"MDPI AG","issue":"17","license":[{"start":{"date-parts":[[2022,8,25]],"date-time":"2022-08-25T00:00:00Z","timestamp":1661385600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Sensors"],"abstract":"<jats:p>Machine learning (ML) is a key technology in smart manufacturing as it provides insights into complex processes without requiring deep domain expertise. This work deals with deep learning algorithms to determine a 3D reconstruction from a single 2D grayscale image. The potential of 3D reconstruction can be used for quality control because the height values contain relevant information that is not visible in 2D data. Instead of 3D scans, estimated depth maps based on a 2D input image can be used with the advantage of a simple setup and a short recording time. Determining a 3D reconstruction from a single input image is a difficult task for which many algorithms and methods have been proposed in the past decades. In this work, three deep learning methods, namely stacked autoencoder (SAE), generative adversarial networks (GANs) and U-Nets are investigated, evaluated and compared for 3D reconstruction from a 2D grayscale image of laser-welded components. In this work, different variants of GANs are tested, with the conclusion that Wasserstein GANs (WGANs) are the most robust approach among them. To the best of our knowledge, the present paper considers for the first time the U-Net, which achieves outstanding results in semantic segmentation, in the context of 3D reconstruction tasks. Unlike the U-Net, which uses standard convolutions, the stacked dilated U-Net (SDU-Net) applies stacked dilated convolutions. Of all the 3D reconstruction approaches considered in this work, the SDU-Net shows the best performance, not only in terms of evaluation metrics but also in terms of computation time. Due to the comparably small number of trainable parameters and the suitability of the architecture for strong data augmentation, a robust model can be generated with only a few training data.<\/jats:p>","DOI":"10.3390\/s22176425","type":"journal-article","created":{"date-parts":[[2022,8,30]],"date-time":"2022-08-30T01:37:55Z","timestamp":1661823475000},"page":"6425","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":19,"title":["Analysis of AI-Based Single-View 3D Reconstruction Methods for an Industrial Application"],"prefix":"10.3390","volume":"22","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-3966-5750","authenticated-orcid":false,"given":"Julia","family":"Hartung","sequence":"first","affiliation":[{"name":"TRUMPF Laser GmbH, Aichhalder Str. 39, 78713 Schramberg, Germany"},{"name":"Institute of Industrial Information Technology, Karlsruhe Institute of Technology, Hertzstra\u00dfe 16, 76187 Karlsruhe, Germany"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4763-9472","authenticated-orcid":false,"given":"Patricia M.","family":"Dold","sequence":"additional","affiliation":[{"name":"TRUMPF Laser GmbH, Aichhalder Str. 39, 78713 Schramberg, Germany"},{"name":"Institute of Industrial Information Technology, Karlsruhe Institute of Technology, Hertzstra\u00dfe 16, 76187 Karlsruhe, Germany"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-0207-0224","authenticated-orcid":false,"given":"Andreas","family":"Jahn","sequence":"additional","affiliation":[{"name":"TRUMPF Laser GmbH, Aichhalder Str. 39, 78713 Schramberg, Germany"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9339-2055","authenticated-orcid":false,"given":"Michael","family":"Heizmann","sequence":"additional","affiliation":[{"name":"Institute of Industrial Information Technology, Karlsruhe Institute of Technology, Hertzstra\u00dfe 16, 76187 Karlsruhe, Germany"}]}],"member":"1968","published-online":{"date-parts":[[2022,8,25]]},"reference":[{"key":"ref_1","unstructured":"Bundesministerium f\u00fcr Wirtschaft und Klimaschutz (BMWK) (2022, January 14). What Is Industrie 4.0?. Available online: https:\/\/www.plattform-i40.de\/IP\/Navigation\/EN\/Industrie40\/WhatIsIndustrie40\/what-is-industrie40.html."},{"key":"ref_2","doi-asserted-by":"crossref","unstructured":"Mayr, A., Meyer, A., Seefried, J., Weigelt, M., Lutz, B., Sultani, D., Hampl, M., and Franke, J. (2017, January 5\u20136). Potentials of machine learning in electric drives production using the example of contacting processes and selective magnet assembly. Proceedings of the IEEE 2017 7th International Electric Drives Production Conference (EDPC), Wuerzburg, Germany.","DOI":"10.1109\/EDPC.2017.8328166"},{"key":"ref_3","doi-asserted-by":"crossref","unstructured":"Weigelt, M., Mayr, A., Seefried, J., Heisler, P., and Franke, J. (2018, January 16\u201319). Conceptual design of an intelligent ultrasonic crimping process using machine learning algorithms. Proceedings of the Procedia Manufacturing, Toyohashi, Japan.","DOI":"10.1016\/j.promfg.2018.10.015"},{"key":"ref_4","doi-asserted-by":"crossref","unstructured":"Vater, J., Pollach, M., Lenz, C., Winkle, D., and Knoll, A. (2021, January 18\u201321). Quality Control and Fault Classification of Laser Welded Hairpins in Electrical Motors. Proceedings of the IEEE 2020 28th European Signal Processing Conference (EUSIPCO), Amsterdam, The Netherlands.","DOI":"10.23919\/Eusipco47968.2020.9287701"},{"key":"ref_5","doi-asserted-by":"crossref","unstructured":"Glaessel, T., Seefried, J., and Franke, J. (2017, January 5\u20136). Challenges in the manufacturing of hairpin windings and application opportunities of infrared lasers for the contacting process. Proceedings of the IEEE 2017 7th International Electric Drives Production Conference (EDPC), Wuerzburg, Germany.","DOI":"10.1109\/EDPC.2017.8328150"},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"2479","DOI":"10.1109\/TIA.2014.2363015","article-title":"Design and Performance of Electrical Propulsion System of Extended Range Electric Vehicle (EREV) Chevrolet Volt","volume":"51","author":"Rahman","year":"2015","journal-title":"IEEE Trans. Ind. Appl."},{"key":"ref_7","doi-asserted-by":"crossref","unstructured":"Jung, D.S., Kim, Y.H., Lee, U.H., and Lee, H.D. (2012, January 6\u20139). Optimum Design of the Electric Vehicle Traction Motor Using the Hairpin Winding. Proceedings of the 2012 IEEE 75th Vehicular Technology Conference (VTC Spring), Yokohama, Japan.","DOI":"10.1109\/VETECS.2012.6240320"},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"67","DOI":"10.1016\/j.neucom.2016.11.062","article-title":"Learning 3D faces from 2D images via Stacked Contractive Autoencoder","volume":"257","author":"Zhang","year":"2017","journal-title":"Neurocomputing"},{"key":"ref_9","doi-asserted-by":"crossref","unstructured":"Baby, A.T., Andrews, A., Dinesh, A., Joseph, A., and Anjusree, V. (2020, January 2\u20134). Face Depth Estimation and 3D Reconstruction. Proceedings of the IEEE 2020 Advanced Computing and Communication Technologies for High Performance Applications (ACCTHPA), Cochin, India.","DOI":"10.1109\/ACCTHPA49271.2020.9213233"},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"23222","DOI":"10.1109\/ACCESS.2019.2898705","article-title":"Face Depth Estimation With Conditional Generative Adversarial Networks","volume":"7","author":"Arslan","year":"2019","journal-title":"IEEE Access"},{"key":"ref_11","doi-asserted-by":"crossref","unstructured":"Isola, P., Zhu, J.Y., Zhou, T., and Efros, A.A. (2017, January 21\u201326). Image-to-Image Translation with Conditional Adversarial Networks. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Honolulu, HI, USA.","DOI":"10.1109\/CVPR.2017.632"},{"key":"ref_12","doi-asserted-by":"crossref","unstructured":"Ronneberger, O., Fischer, P., and Brox, T. (2015). U-net: Convolutional networks for biomedical image segmentation. arXiv.","DOI":"10.1007\/978-3-319-24574-4_28"},{"key":"ref_13","unstructured":"Hartung, J., Jahn, A., Stambke, M., Wehner, O., Thieringer, R., and Heizmann, M. (2020, January 26\u201327). Camera-based spatter detection in laser welding with a deep learning approach. Proceedings of the Forum Bildverarbeitung 2020, Karlsruhe, Germany."},{"key":"ref_14","unstructured":"Wang, S., Hu, S.Y., Cheah, E., Wang, X., Wang, J., Chen, L., Baikpour, M., Ozturk, A., Li, Q., and Chou, S.H. (2020). U-Net using stacked dilated convolutions for medical image segmentation. arXiv."},{"key":"ref_15","doi-asserted-by":"crossref","unstructured":"Hartung, J., Jahn, A., Bocksrocker, O., and Heizmann, M. (2021). Camera-based in-process quality measurement of hairpin welding. Appl. Sci., 11.","DOI":"10.3390\/app112110375"},{"key":"ref_16","doi-asserted-by":"crossref","unstructured":"Yaning, Y., Zhang, Z., Zhang, W., Zhang, C., Li, W., and Zhao, T. (2019). Semantic Segmentation of Urban Buildings from VHR Remote Sensing Imagery Using a Deep Convolutional Neural Network. Remote Sens., 11.","DOI":"10.3390\/rs11151774"},{"key":"ref_17","unstructured":"Horn, B.K.P., and Brooks, M.J. (1989). Shape from Shading, MIT Press."},{"key":"ref_18","doi-asserted-by":"crossref","unstructured":"Schonberger, J.L., and Frahm, J.M. (July, January 26). Structure-from-Motion Revisited. Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition, Las Vegas, NV, USA.","DOI":"10.1109\/CVPR.2016.445"},{"key":"ref_19","doi-asserted-by":"crossref","unstructured":"Seitz, S., Curless, B., Diebel, J., Scharstein, D., and Szeliski, R. (2006, January 17\u201322). A Comparison and Evaluation of Multi-View Stereo Reconstruction Algorithms. Proceedings of the 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition\u2014Volume 1, New York, NY, USA.","DOI":"10.1109\/CVPR.2006.19"},{"key":"ref_20","doi-asserted-by":"crossref","unstructured":"Miyazaki, D., Tan, R.T., Hara, K., and Ikeuchi, K. (2003, January 13\u201316). Polarization-based inverse rendering from a single view. Proceedings of the IEEE International Conference on Computer Vision, Nice, France.","DOI":"10.1109\/ICCV.2003.1238455"},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"1653","DOI":"10.1109\/TIP.2006.871114","article-title":"Recovery of surface orientation from diffuse polarization","volume":"15","author":"Atkinson","year":"2006","journal-title":"IEEE Trans. Image Process."},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"64","DOI":"10.1007\/s11263-012-0546-3","article-title":"Shape and refractive index from single-view spectro-polarimetric images","volume":"101","author":"Huynh","year":"2013","journal-title":"Int. J. Comput. Vis."},{"key":"ref_23","unstructured":"Miyazaki, D., Kagesawa, M., and Ikeuchi, K. (2002, January 11\u201313). Determining Shapes of Transparent Objects from Two Polarization Images. Proceedings of the IAPR Workshop on Machine Vision Applications, Nara, Japan."},{"key":"ref_24","unstructured":"Rahmann, S., and Canterakis, N. (2001, January 8\u201314). Reconstruction of specular surfaces using polarization imaging. Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. CVPR 2001, Kauai, HI, USA."},{"key":"ref_25","doi-asserted-by":"crossref","first-page":"11","DOI":"10.1134\/S0030400X06070034","article-title":"Polarization imaging for 3D inspection of highly reflective metallic objects","volume":"101","author":"Morel","year":"2006","journal-title":"Opt. Spectrosc."},{"key":"ref_26","doi-asserted-by":"crossref","unstructured":"Soltani, A.A., Huang, H., Wu, J., Kulkarni, T.D., and Tenenbaum, J.B. (2017, January 21\u201326). Synthesizing 3D Shapes via Modeling Multi-view Depth Maps and Silhouettes with Deep Generative Networks. Proceedings of the 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, HI, USA.","DOI":"10.1109\/CVPR.2017.269"},{"key":"ref_27","doi-asserted-by":"crossref","unstructured":"Choy, C.B., Xu, D., Gwak, J., Chen, K., and Savarese, S. (2016, January 11\u201314). 3D-R2N2: A Unified Approach for Single and Multi-view 3D Object Reconstruction. Proceedings of the European Conference on Computer Vision, Amsterdam, The Netherlands.","DOI":"10.1007\/978-3-319-46484-8_38"},{"key":"ref_28","doi-asserted-by":"crossref","first-page":"2339","DOI":"10.1109\/LSP.2015.2480802","article-title":"DeepPano: Deep Panoramic Representation for 3-D Shape Recognition","volume":"22","author":"Shi","year":"2015","journal-title":"IEEE Signal Process. Lett."},{"key":"ref_29","doi-asserted-by":"crossref","unstructured":"Liu, F., Shen, C., and Lin, G. (2014, January 23\u201328). Deep Convolutional Neural Fields for Depth Estimation from a Single Image. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Columbus, OH, USA.","DOI":"10.1109\/CVPR.2015.7299152"},{"key":"ref_30","doi-asserted-by":"crossref","first-page":"824","DOI":"10.1109\/TPAMI.2008.132","article-title":"Make3D: Learning 3D Scene Structure from a Single Still Image","volume":"31","author":"Saxena","year":"2009","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"ref_31","unstructured":"Goodfellow, I.J., Pouget-Abadie, J., Mirza, M., Xu, B., Warde-Farley, D., Ozair, S., Courville, A., and Bengio, Y. (2014). Generative Adversarial Networks. arXiv."},{"key":"ref_32","unstructured":"Mirza, M., and Osindero, S. (2014). Conditional Generative Adversarial Nets. arXiv."},{"key":"ref_33","doi-asserted-by":"crossref","first-page":"14048","DOI":"10.1109\/ACCESS.2018.2808938","article-title":"cC-GAN: A Robust Transfer-Learning Framework for HEp-2 Specimen Image Segmentation","volume":"6","author":"Li","year":"2018","journal-title":"IEEE Access"},{"key":"ref_34","doi-asserted-by":"crossref","unstructured":"Ledig, C., Theis, L., Huszar, F., Caballero, J., Cunningham, A., Acosta, A., Aitken, A., Tejani, A., Totz, J., and Wang, Z. (2017, January 21\u201326). Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Honolulu, HI, USA.","DOI":"10.1109\/CVPR.2017.19"},{"key":"ref_35","unstructured":"Arjovsky, M., and Bottou, L. (2017). Towards Principled Methods for Training Generative Adversarial Networks. arXiv."},{"key":"ref_36","doi-asserted-by":"crossref","unstructured":"Mao, X., Li, Q., Xie, H., Lau, R.Y.K., Wang, Z., and Smolley, S.P. (2017, January 22\u201329). Least Squares Generative Adversarial Networks. Proceedings of the IEEE International Conference on Computer Vision, Venice, Italy.","DOI":"10.1109\/ICCV.2017.304"},{"key":"ref_37","unstructured":"Arjovsky, M., Chintala, S., and Bottou, L. (2017). Wasserstein GAN. arXiv."},{"key":"ref_38","doi-asserted-by":"crossref","unstructured":"Huang, X., Li, Y., Poursaeed, O., Hopcroft, J., and Belongie, S. (2017, January 21\u201326). Stacked Generative Adversarial Networks. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Honolulu, HI, USA.","DOI":"10.1109\/CVPR.2017.202"},{"key":"ref_39","unstructured":"Zhang, H., Goodfellow, I., Metaxas, D., and Odena, A. (2019, January 24). Self-Attention Generative Adversarial Networks. Proceedings of the International Conference on Machine Learning, Silchar, India."},{"key":"ref_40","doi-asserted-by":"crossref","unstructured":"Arslan, A.T., and Seke, E. (2019, January 11\u201313). Training Wasserstein GANs for Estimating Depth Maps. Proceedings of the IEEE 2019 3rd International Symposium on Multidisciplinary Studies and Innovative Technologies (ISMSIT), Ankara, Turkey.","DOI":"10.1109\/ISMSIT.2019.8932868"},{"key":"ref_41","doi-asserted-by":"crossref","unstructured":"Long, J., Shelhamer, E., and Darrell, T. (2015, January 7\u201312). Fully Convolutional Networks for Semantic Segmentation. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Boston, MA, USA.","DOI":"10.1109\/CVPR.2015.7298965"},{"key":"ref_42","unstructured":"Oktay, O., Schlemper, J., Folgoc, L.L., Lee, M., Heinrich, M., Misawa, K., Mori, K., McDonagh, S., Hammerla, N.Y., and Kainz, B. (2018). Attention U-Net: Learning Where to Look for the Pancreas. arXiv."},{"key":"ref_43","doi-asserted-by":"crossref","unstructured":"Alom, M.Z., Hasan, M., Yakopcic, C., Taha, T.M., and Asari, V.K. (2018). Recurrent Residual Convolutional Neural Network based on U-Net (R2U-Net) for Medical Image Segmentation. arXiv.","DOI":"10.1109\/NAECON.2018.8556686"},{"key":"ref_44","doi-asserted-by":"crossref","unstructured":"Zhou, Z., Siddiquee, M.M.R., Tajbakhsh, N., and Liang, J. (2018). UNet++: A Nested U-Net Architecture for Medical Image Segmentation. Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support, Springer.","DOI":"10.1007\/978-3-030-00889-5_1"},{"key":"ref_45","doi-asserted-by":"crossref","first-page":"3244","DOI":"10.1364\/BOE.9.003244","article-title":"DRUNET: A Dilated-Residual U-Net Deep Learning Network to Digitally Stain Optic Nerve Head Tissues in Optical Coherence Tomography Images","volume":"9","author":"Devalla","year":"2018","journal-title":"Biomed. Opt. Express"},{"key":"ref_46","doi-asserted-by":"crossref","unstructured":"Hamaguchi, R., Fujita, A., Nemoto, K., Imaizumi, T., and Hikosaka, S. (2018, January 12\u201315). Effective Use of Dilated Convolutions for Segmenting Small Object Instances in Remote Sensing Imagery. Proceedings of the 2018 IEEE Winter Conference on Applications of Computer Vision (WACV), Lake Tahoe, NV, USA.","DOI":"10.1109\/WACV.2018.00162"},{"key":"ref_47","doi-asserted-by":"crossref","first-page":"1825","DOI":"10.1007\/s00170-017-0461-4","article-title":"Quality assessment in laser welding: A critical review","volume":"94","author":"Stavridis","year":"2018","journal-title":"Int. J. Adv. Manuf. Technol."},{"key":"ref_48","doi-asserted-by":"crossref","unstructured":"D\u00f6ssel, O. (2016). Bildgebende Verfahren in der Medizin\u2014Von der Technik zur Medizinischen Anwendung, Springer.","DOI":"10.1007\/978-3-642-54407-1"},{"key":"ref_49","unstructured":"Ghahramani, Z., Welling, M., Cortes, C., Lawrence, N., and Weinberger, K. Discriminative Unsupervised Feature Learning with Convolutional Neural Networks. Proceedings of the Advances in Neural Information Processing Systems."},{"key":"ref_50","unstructured":"Radford, A., Metz, L., and Chintala, S. (2015). Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks. arXiv."},{"key":"ref_51","doi-asserted-by":"crossref","first-page":"759","DOI":"10.1007\/s10845-019-01476-x","article-title":"Segmentation-based deep-learning approach for surface-defect detection","volume":"31","author":"Tabernik","year":"2019","journal-title":"J. Intell. Manuf."}],"container-title":["Sensors"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1424-8220\/22\/17\/6425\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,11]],"date-time":"2025-10-11T00:15:33Z","timestamp":1760141733000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1424-8220\/22\/17\/6425"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,8,25]]},"references-count":51,"journal-issue":{"issue":"17","published-online":{"date-parts":[[2022,9]]}},"alternative-id":["s22176425"],"URL":"https:\/\/doi.org\/10.3390\/s22176425","relation":{},"ISSN":["1424-8220"],"issn-type":[{"value":"1424-8220","type":"electronic"}],"subject":[],"published":{"date-parts":[[2022,8,25]]}}}