{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,11,4]],"date-time":"2025-11-04T11:03:12Z","timestamp":1762254192323,"version":"build-2065373602"},"reference-count":50,"publisher":"MDPI AG","issue":"14","license":[{"start":{"date-parts":[[2022,7,16]],"date-time":"2022-07-16T00:00:00Z","timestamp":1657929600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100004329","name":"Slovenian Research Agency (ARRS)","doi-asserted-by":"publisher","award":["J2-9228 (B)","P2-0214 (B)"],"award-info":[{"award-number":["J2-9228 (B)","P2-0214 (B)"]}],"id":[{"id":"10.13039\/501100004329","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Sensors"],"abstract":"<jats:p>The task of reconstructing 3D scenes based on visual data represents a longstanding problem in computer vision. Common reconstruction approaches rely on the use of multiple volumetric primitives to describe complex objects. Superquadrics (a class of volumetric primitives) have shown great promise due to their ability to describe various shapes with only a few parameters. Recent research has shown that deep learning methods can be used to accurately reconstruct random superquadrics from both 3D point cloud data and simple depth images. In this paper, we extended these reconstruction methods to intensity and color images. Specifically, we used a dedicated convolutional neural network (CNN) model to reconstruct a single superquadric from the given input image. We analyzed the results in a qualitative and quantitative manner, by visualizing reconstructed superquadrics as well as observing error and accuracy distributions of predictions. We showed that a CNN model designed around a simple ResNet backbone can be used to accurately reconstruct superquadrics from images containing one object, but only if one of the spatial parameters is fixed or if it can be determined from other image characteristics, e.g., shadows. Furthermore, we experimented with images of increasing complexity, for example, by adding textures, and observed that the results degraded only slightly. In addition, we show that our model outperforms the current state-of-the-art method on the studied task. Our final result is a highly accurate superquadric reconstruction model, which can also reconstruct superquadrics from real images of simple objects, without additional training.<\/jats:p>","DOI":"10.3390\/s22145332","type":"journal-article","created":{"date-parts":[[2022,7,18]],"date-time":"2022-07-18T01:53:22Z","timestamp":1658109202000},"page":"5332","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":3,"title":["Reconstructing Superquadrics from Intensity and Color Images"],"prefix":"10.3390","volume":"22","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-3463-7670","authenticated-orcid":false,"given":"Darian","family":"Toma\u0161evi\u0107","sequence":"first","affiliation":[{"name":"Faculty of Computer and Information Science, University of Ljubljana, 1000 Ljubljana, Slovenia"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9744-4035","authenticated-orcid":false,"given":"Peter","family":"Peer","sequence":"additional","affiliation":[{"name":"Faculty of Computer and Information Science, University of Ljubljana, 1000 Ljubljana, Slovenia"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9268-6825","authenticated-orcid":false,"given":"Franc","family":"Solina","sequence":"additional","affiliation":[{"name":"Faculty of Computer and Information Science, University of Ljubljana, 1000 Ljubljana, Slovenia"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8765-897X","authenticated-orcid":false,"given":"Ale\u0161","family":"Jakli\u010d","sequence":"additional","affiliation":[{"name":"Faculty of Computer and Information Science, University of Ljubljana, 1000 Ljubljana, Slovenia"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3385-5780","authenticated-orcid":false,"given":"Vitomir","family":"\u0160truc","sequence":"additional","affiliation":[{"name":"Faculty of Electrical Engineering, University of Ljubljana, 1000 Ljubljana, Slovenia"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1968","published-online":{"date-parts":[[2022,7,16]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"11","DOI":"10.1109\/MCG.1981.1673799","article-title":"Superquadrics and angle-preserving transformations","volume":"1","author":"Barr","year":"1981","journal-title":"IEEE Comput. Graph. Appl."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"131","DOI":"10.1109\/34.44401","article-title":"Recovery of parametric models from range images: The case for superquadrics with global deformations","volume":"12","author":"Solina","year":"1990","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"ref_3","unstructured":"Khosla, P., and Volpe, R. (1988, January 24\u201329). Superquadric artificial potentials for obstacle avoidance and approach. Proceedings of the IEEE International Conference on Robotics and Automation, Philadelphia, PA, USA."},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"65","DOI":"10.2514\/1.D0170","article-title":"Incorporating stochastics into optimal collision avoidance problems using superquadrics","volume":"28","author":"Smith","year":"2020","journal-title":"J. Air Transp."},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"eaau4984","DOI":"10.1126\/scirobotics.aau4984","article-title":"Learning ambidextrous robot grasping policies","volume":"4","author":"Mahler","year":"2019","journal-title":"Sci. Robot."},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"183","DOI":"10.1177\/0278364919859066","article-title":"Learning robust, real-time, reactive robotic grasping","volume":"39","author":"Morrison","year":"2020","journal-title":"Int. J. Robot. Res."},{"key":"ref_7","unstructured":"Marr, D. (1982). Vision: A Computational Investigation into the Human Representation and Processing of Visual Information, MIT Press."},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"293","DOI":"10.1016\/0004-3702(86)90052-4","article-title":"Perceptual organization and the representation of natural form","volume":"28","author":"Pentland","year":"1986","journal-title":"Artif. Intell."},{"key":"ref_9","doi-asserted-by":"crossref","unstructured":"Tulsiani, S., Su, H., Guibas, L.J., Efros, A.A., and Malik, J. (2017, January 21\u201326). Learning shape abstractions by assembling volumetric primitives. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Honolulu, HI, USA.","DOI":"10.1109\/CVPR.2017.160"},{"key":"ref_10","doi-asserted-by":"crossref","unstructured":"Paschalidou, D., Ulusoy, A.O., and Geiger, A. (2019, January 15\u201320). Superquadrics revisited: Learning 3D shape parsing beyond cuboids. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, Long Beach, CA, USA.","DOI":"10.1109\/CVPR.2019.01059"},{"key":"ref_11","doi-asserted-by":"crossref","unstructured":"Paschalidou, D., Gool, L.V., and Geiger, A. (2020, January 13\u201319). Learning unsupervised hierarchical part decomposition of 3D objects from a single RGB image. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, Seattle, WA, USA.","DOI":"10.1109\/CVPR42600.2020.00114"},{"key":"ref_12","doi-asserted-by":"crossref","unstructured":"Oblak, T., Grm, K., Jakli\u010d, A., Peer, P., \u0160truc, V., and Solina, F. (2019, January 3\u20135). Recovery of superquadrics from range images using deep learning: A preliminary study. Proceedings of the IEEE International Work Conference on Bioinspired Intelligence (IWOBI), Budapest, Hungary.","DOI":"10.1109\/IWOBI47054.2019.9114452"},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"1087","DOI":"10.1109\/ACCESS.2020.3041584","article-title":"Learning to Predict Superquadric Parameters From Depth Images With Explicit and Implicit Supervision","volume":"9","author":"Oblak","year":"2020","journal-title":"IEEE Access"},{"key":"ref_14","unstructured":"\u0160ircelj, J., Oblak, T., Grm, K., Petkovi\u0107, U., Jakli\u010d, A., Peer, P., \u0160truc, V., and Solina, F. (2020, January 3\u20135). Segmentation and recovery of superquadric models using convolutional neural networks. Proceedings of the 25th Computer Vision Winter Workshop, Roga\u0161ka Slatina, Slovenia."},{"key":"ref_15","doi-asserted-by":"crossref","unstructured":"Li, S., Liu, M., and Walder, C. (2021). EditVAE: Unsupervised Part-Aware Controllable 3D Point Cloud Shape Generation. arXiv.","DOI":"10.1609\/aaai.v36i2.20027"},{"key":"ref_16","doi-asserted-by":"crossref","unstructured":"Abrams, A., Miskell, K., and Pless, R. (2013, January 23\u201328). The episolar constraint: Monocular shape from shadow correspondence. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Portland, OR, USA.","DOI":"10.1109\/CVPR.2013.185"},{"key":"ref_17","doi-asserted-by":"crossref","unstructured":"Xie, Y., Feng, D., Xiong, S., Zhu, J., and Liu, Y. (2021). Multi-Scene Building Height Estimation Method Based on Shadow in High Resolution Imagery. Remote Sens., 13.","DOI":"10.3390\/rs13152862"},{"key":"ref_18","unstructured":"Vezzani, G., Pattacini, U., and Natale, L. (June, January 29). A grasping approach based on superquadric models. Proceedings of the IEEE International Conference on Robotics and Automation (ICRA), Singapore."},{"key":"ref_19","unstructured":"Makhal, A., Thomas, F., and Gracia, A.P. (February, January 31). Grasping unknown objects in clutter by superquadric representation. Proceedings of the 2nd IEEE International Conference on Robotic Computing (IRC), Laguna Hills, CA, USA."},{"key":"ref_20","doi-asserted-by":"crossref","unstructured":"Vezzani, G., Pattacini, U., Pasquale, G., and Natale, L. (2018, January 21\u201325). Improving Superquadric Modeling and Grasping with Prior on Object Shapes. Proceedings of the IEEE International Conference on Robotics and Automation (ICRA), Brisbane, Australia.","DOI":"10.1109\/ICRA.2018.8463161"},{"key":"ref_21","unstructured":"Haschke, R., Walck, G., and Ritter, H. (October, January 27). Geometry-Based Grasping Pipeline for Bi-Modal Pick and Place. Proceedings of the IEEE\/RSJ International Conference on Intelligent Robots and Systems (IROS), Prague, Czech Republic."},{"key":"ref_22","unstructured":"Solina, F., and Bajcsy, R. (1987, January 13\u201317). Range image interpretation of mail pieces with superquadrics. Proceedings of the National Conference on Artificial Intelligence, Seattle, WA, USA."},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"143","DOI":"10.1016\/j.jas.2015.08.007","article-title":"Volumetric models from 3D point clouds: The case study of sarcophagi cargo from a 2nd\/3rd century AD Roman shipwreck near Sutivan on island Bra\u010d, Croatia","volume":"62","author":"Solina","year":"2015","journal-title":"J. Archaeol. Sci."},{"key":"ref_24","unstructured":"Munih, M. (2017). 3D modeliranje podvodnih posnetkov. SI Robotika, Slovenska Matica."},{"key":"ref_25","doi-asserted-by":"crossref","unstructured":"Hachiuma, R., and Saito, H. (2019, January 23\u201325). Volumetric Representation of Semantically Segmented Human Body Parts Using Superquadrics. Proceedings of the International Conference on Virtual Reality and Augmented Reality, Tallinn, Estonia.","DOI":"10.1109\/ISMAR-Adjunct.2019.00-50"},{"key":"ref_26","unstructured":"Pentland, A.P. (1987, January 8\u201311). Recognition by parts. Proceedings of the IEEE 1st International Conference on Computer Vision, London, UK."},{"key":"ref_27","doi-asserted-by":"crossref","first-page":"358","DOI":"10.1117\/12.942759","article-title":"Recovery of superquadrics from 3D information","volume":"Volume 848","author":"Boult","year":"1988","journal-title":"Proceedings of the Intelligent Robots and Computer Vision VI"},{"key":"ref_28","unstructured":"Gross, A.D., and Boult, T.E. (1988, January 5\u20138). Error of fit measures for recovering parametric solids. Proceedings of the 2nd International Conference of Computer Vision, Tampa, FL, USA."},{"key":"ref_29","doi-asserted-by":"crossref","first-page":"771","DOI":"10.1109\/34.236252","article-title":"Darboux frames, snakes, and super-quadrics: Geometry from the bottom up","volume":"15","author":"Ferrie","year":"1993","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"ref_30","doi-asserted-by":"crossref","first-page":"191","DOI":"10.1016\/S0734-189X(88)80005-7","article-title":"Hyperquadrics: Smoothly deformable shapes with convex polyhedral bounds","volume":"44","author":"Hanson","year":"1988","journal-title":"Comput. Vision Graph. Image Process."},{"key":"ref_31","doi-asserted-by":"crossref","first-page":"703","DOI":"10.1109\/34.85659","article-title":"Dynamic 3D models with local and global deformations: Deformable superquadrics","volume":"13","author":"Terzopoulos","year":"1991","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"ref_32","doi-asserted-by":"crossref","first-page":"1289","DOI":"10.1109\/34.632988","article-title":"Superquadrics for segmenting and modeling range data","volume":"19","author":"Leonardis","year":"1997","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"ref_33","doi-asserted-by":"crossref","first-page":"105","DOI":"10.1016\/j.cviu.2003.11.002","article-title":"Part-level object recognition using superquadrics","volume":"95","author":"Krivic","year":"2004","journal-title":"Comput. Vis. Image Underst."},{"key":"ref_34","doi-asserted-by":"crossref","unstructured":"Slabanja, J., Meden, B., Peer, P., Jakli\u010d, A., and Solina, F. (2018, January 17\u201319). Segmentation and reconstruction of 3D models from a point cloud with deep neural networks. Proceedings of the International Conference on Information and Communication Technology Convergence (ICTC), Jeju Island, Korea.","DOI":"10.1109\/ICTC.2018.8539500"},{"key":"ref_35","unstructured":"Wu, Z., Song, S., Khosla, A., Yu, F., Zhang, L., Tang, X., and Xiao, J. (2015, January 7\u201312). 3D shapenets: A deep representation for volumetric shapes. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Boston, MA, USA."},{"key":"ref_36","first-page":"540","article-title":"MarrNet: 3D Shape Reconstruction via 2.5D Sketches","volume":"30","author":"Wu","year":"2017","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_37","doi-asserted-by":"crossref","first-page":"1352","DOI":"10.1109\/TMI.2016.2521800","article-title":"A CNN regression approach for real-time 2D\/3D registration","volume":"35","author":"Miao","year":"2016","journal-title":"IEEE Trans. Med. Imaging"},{"key":"ref_38","doi-asserted-by":"crossref","unstructured":"Zhu, R., Kiani Galoogahi, H., Wang, C., and Lucey, S. (2017, January 22\u201329). Rethinking reprojection: Closing the loop for pose-aware shape reconstruction from a single image. Proceedings of the IEEE International Conference on Computer Vision, Venice, Italy.","DOI":"10.1109\/ICCV.2017.16"},{"key":"ref_39","doi-asserted-by":"crossref","unstructured":"Xiang, Y., Schmidt, T., Narayanan, V., and Fox, D. (2018, January 26\u201330). PoseCNN: A Convolutional Neural Network for 6D Object Pose Estimation in Cluttered Scenes. Proceedings of the 14th Robotics: Science and Systems (RSS), Pittsburgh, PA, USA.","DOI":"10.15607\/RSS.2018.XIV.019"},{"key":"ref_40","doi-asserted-by":"crossref","unstructured":"Kuipers, J.B. (1999). Quaternions and Rotation Sequences: A Primer with Applications to Orbits, Aerospace, and Virtual Reality, Princeton University Press.","DOI":"10.1515\/9780691211701"},{"key":"ref_41","doi-asserted-by":"crossref","unstructured":"Jakli\u010d, A., Leonardis, A., and Solina, F. (2000). Segmentation and Recovery of Superquadrics, Springer Science & Business Media.","DOI":"10.1007\/978-94-015-9456-1"},{"key":"ref_42","doi-asserted-by":"crossref","unstructured":"He, K., Zhang, X., Ren, S., and Sun, J. (2016, January 27\u201330). Deep residual learning for image recognition. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Las Vegas, NV, USA.","DOI":"10.1109\/CVPR.2016.90"},{"key":"ref_43","doi-asserted-by":"crossref","unstructured":"Shoemake, K. (1992). Uniform random rotations. Graphics Gems III (IBM Version), Elsevier.","DOI":"10.1016\/B978-0-08-050755-2.50036-1"},{"key":"ref_44","unstructured":"Oechsle, M., Mescheder, L., Niemeyer, M., Strauss, T., and Geiger, A. (November, January 27). Texture fields: Learning texture representations in function space. Proceedings of the IEEE\/CVF International Conference on Computer Vision, Seoul, Korea."},{"key":"ref_45","doi-asserted-by":"crossref","unstructured":"Deng, J., Dong, W., Socher, R., Li, L.J., Li, K., and Fei-Fei, L. (2009, January 20\u201325). ImageNet: A large-scale hierarchical image database. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Miami, FL, USA.","DOI":"10.1109\/CVPR.2009.5206848"},{"key":"ref_46","unstructured":"Kingma, D.P., and Ba, J.L. (2015, January 7\u20139). Adam: A method for stochastic optimization. Proceedings of the International Conference on Learning Representations, San Diego, CA, USA."},{"key":"ref_47","doi-asserted-by":"crossref","unstructured":"Bengio, Y. (2012). Practical recommendations for gradient-based training of deep architectures. Neural Networks: Tricks of the Trade, Springer.","DOI":"10.1007\/978-3-642-35289-8_26"},{"key":"ref_48","unstructured":"Keskar, N.S., Nocedal, J., Tang, P.T.P., Mudigere, D., and Smelyanskiy, M. (2017, January 24\u201326). On large-batch training for deep learning: Generalization gap and sharp minima. Proceedings of the 5th International Conference on Learning Representations, Toulon, France."},{"key":"ref_49","doi-asserted-by":"crossref","first-page":"312","DOI":"10.1016\/j.icte.2020.04.010","article-title":"The effect of batch size on the generalizability of the convolutional neural networks on a histopathology dataset","volume":"6","author":"Kandel","year":"2020","journal-title":"ICT Express"},{"key":"ref_50","doi-asserted-by":"crossref","unstructured":"Szegedy, C., Vanhoucke, V., Ioffe, S., Shlens, J., and Wojna, Z. (2016, January 27\u201330). Rethinking the inception architecture for computer vision. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Las Vegas, NV, USA.","DOI":"10.1109\/CVPR.2016.308"}],"container-title":["Sensors"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1424-8220\/22\/14\/5332\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,10]],"date-time":"2025-10-10T23:52:13Z","timestamp":1760140333000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1424-8220\/22\/14\/5332"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,7,16]]},"references-count":50,"journal-issue":{"issue":"14","published-online":{"date-parts":[[2022,7]]}},"alternative-id":["s22145332"],"URL":"https:\/\/doi.org\/10.3390\/s22145332","relation":{},"ISSN":["1424-8220"],"issn-type":[{"type":"electronic","value":"1424-8220"}],"subject":[],"published":{"date-parts":[[2022,7,16]]}}}