{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,3,24]],"date-time":"2025-03-24T08:01:43Z","timestamp":1742803303938},"reference-count":42,"publisher":"Wiley","issue":"2","license":[{"start":{"date-parts":[[2018,5,22]],"date-time":"2018-05-22T00:00:00Z","timestamp":1526947200000},"content-version":"vor","delay-in-days":21,"URL":"http:\/\/onlinelibrary.wiley.com\/termsAndConditions#vor"}],"content-domain":{"domain":["onlinelibrary.wiley.com"],"crossmark-restriction":true},"short-container-title":["Computer Graphics Forum"],"published-print":{"date-parts":[[2018,5]]},"abstract":"Abstract<\/jats:title>As many different 3D volumes could produce the same 2D x\u2010ray image, inverting this process is challenging. We show that recent deep learning\u2010based convolutional neural networks can solve this task. As the main challenge in learning is the sheer amount of data created when extending the 2D image into a 3D volume, we suggest firstly to learn a coarse, fixed\u2010resolution volume which is then fused in a second step with the input x\u2010ray into a high\u2010resolution volume. To train and validate our approach we introduce a new dataset that comprises of close to half a million computer\u2010simulated 2D x\u2010ray images of 3D volumes scanned from 175 mammalian species. Future applications of our approach include stereoscopic rendering of legacy x\u2010ray images, re\u2010rendering of x\u2010rays including changes of illumination, view pose or geometry. Our evaluation includes comparison to previous tomography work, previous learning methods using our data, a user study and application to a set of real x\u2010rays.<\/jats:p>","DOI":"10.1111\/cgf.13369","type":"journal-article","created":{"date-parts":[[2018,6,1]],"date-time":"2018-06-01T00:19:39Z","timestamp":1527812379000},"page":"377-388","update-policy":"http:\/\/dx.doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":79,"title":["Single\u2010image Tomography: 3D Volumes from 2D Cranial X\u2010Rays"],"prefix":"10.1111","volume":"37","author":[{"given":"Phlipp","family":"Henzler","sequence":"first","affiliation":[{"name":"Ulm University Germany"}]},{"given":"Volker","family":"Rasche","sequence":"additional","affiliation":[{"name":"Ulm University Germany"}]},{"given":"Timo","family":"Ropinski","sequence":"additional","affiliation":[{"name":"Ulm University Germany"}]},{"given":"Tobias","family":"Ritschel","sequence":"additional","affiliation":[{"name":"University College London UK"}]}],"member":"311","published-online":{"date-parts":[[2018,5,22]]},"reference":[{"key":"e_1_2_8_2_2","doi-asserted-by":"crossref","unstructured":"BahramiK. ShiF. RekikI. ShenD.: Convolutional neural network for reconstruction of 7T\u2010like images from 3T MRI using appearance and anatomical features. InLarge\u2010Scale Annotation of Biomedical Data and Expert Label Synthesis(2016) pp.39\u201347. 3","DOI":"10.1007\/978-3-319-46976-8_5"},{"key":"e_1_2_8_3_2","doi-asserted-by":"crossref","unstructured":"\u00c7i\u00e7ek\u00d6. AbdulkadirA. LienkampS. S. BroxT. RonnebergerO.: 3D U\u2010Net: learning dense volumetric segmentation from sparse annotation. InProc. Medical Image Computing and Computer\u2010Assisted Intervention(2016) pp.424\u201332. 5","DOI":"10.1007\/978-3-319-46723-8_49"},{"key":"e_1_2_8_4_2","doi-asserted-by":"crossref","unstructured":"ChoyC. B. XuD. GwakJ. ChenK. SavareseS.: 3d\u2010r2n2: A unified approach for single and multi\u2010view 3d object reconstruction. InECCV(2016) pp.628\u201344. 2","DOI":"10.1007\/978-3-319-46484-8_38"},{"key":"e_1_2_8_5_2","doi-asserted-by":"crossref","unstructured":"DrebinR. A. CarpenterL. HanrahanP.: Volume rendering. InSiggraph Computer Graphics(1988) Vol.22 pp.65\u201374. 2","DOI":"10.1145\/378456.378484"},{"key":"e_1_2_8_6_2","doi-asserted-by":"crossref","unstructured":"DosovitskiyA. FischerP. IlgE. HausserP. HazirbasC. GolkovV. van der SmagtP. C remersD. BroxT.: Flownet: Learning optical flow with convolutional networks. InProc. ICCV(2015) pp.2758\u201366. 3","DOI":"10.1109\/ICCV.2015.316"},{"key":"e_1_2_8_7_2","unstructured":"Digmorph:Digimorph 2017. 3 8"},{"key":"e_1_2_8_8_2","volume-title":"Encyclopedia of Optical Engineering","author":"Driggers R. G.","year":"2003"},{"key":"e_1_2_8_9_2","doi-asserted-by":"publisher","DOI":"10.1201\/b10629"},{"key":"e_1_2_8_10_2","unstructured":"EigenD. PuhrschC. FergusR.: Depth map prediction from a single image using a multi\u2010scale deep network. InNIPS(2014) pp.2366\u201374. 2"},{"key":"e_1_2_8_11_2","doi-asserted-by":"crossref","unstructured":"FirmanM. Mac AodhaO. JulierS. BrostowG. J.: Structured prediction of unobserved voxels from a single depth image. InCVPR(2016). 2","DOI":"10.1109\/CVPR.2016.586"},{"key":"e_1_2_8_12_2","doi-asserted-by":"crossref","unstructured":"FanH. SuH. GuibasL.: A point set generation network for 3d object reconstruction from a single image.arXiv:1612.00603(2016). 2","DOI":"10.1109\/CVPR.2017.264"},{"key":"e_1_2_8_13_2","doi-asserted-by":"crossref","unstructured":"GirdharR. FouheyD. F. RodriguezM. GuptaA.: Learning a predictable and generative vector representation for objects. InECCV(2016) pp.484\u201399. 2","DOI":"10.1007\/978-3-319-46466-4_29"},{"key":"e_1_2_8_14_2","doi-asserted-by":"publisher","DOI":"10.1259\/0007-1285-46-552-1016"},{"key":"e_1_2_8_15_2","doi-asserted-by":"crossref","unstructured":"HammernikK. W\u00fcrflT. PockT. MaierA.: A deep learning architecture for limited\u2010angle computed tomography reconstruction. InBildverarbeitung f\u00fcr die Medizin2017.2017 pp.92\u20137. 3","DOI":"10.1007\/978-3-662-54345-0_25"},{"key":"e_1_2_8_16_2","unstructured":"HeK. ZhangX. RenS. SunJ.: Deep residual learning for image recognition. InCVPR(2016) pp.770\u20138. 4 5"},{"key":"e_1_2_8_17_2","doi-asserted-by":"crossref","unstructured":"IhrkeI. MagnorM.: Image\u2010based tomographic reconstruction of flames. InProc. SCA(2004) pp.365\u201373. 2","DOI":"10.1145\/1028523.1028572"},{"key":"e_1_2_8_18_2","unstructured":"IsolaP. ZhuJ.\u2010Y. ZhouT. EfrosA. A.: Image\u2010to\u2010image translation with conditional adversarial networks.arXiv:1611.07004(2016). 10"},{"key":"e_1_2_8_19_2","unstructured":"JiaY. ShelhamerE. DonahueJ. KarayevS. LongJ. GirshickR. GuadarramaS. DarrellT.: Caffe: Convolutional architecture for fast feature embedding. InProc. ACM Multimedia(2014) pp.675\u2013678. 2 5"},{"key":"e_1_2_8_20_2","doi-asserted-by":"publisher","DOI":"10.1038\/nature14539"},{"key":"e_1_2_8_21_2","doi-asserted-by":"publisher","DOI":"10.1002\/mrm.21391"},{"key":"e_1_2_8_22_2","unstructured":"LongJ. ShelhamerE. DarrellT.: Fully convolutional networks for semantic segmentation. InCVPR(2015) pp.3431\u201340. 4 5"},{"key":"e_1_2_8_23_2","doi-asserted-by":"crossref","unstructured":"LameckerH. WenckebachT. H. HegeH.\u2010C.: Atlasbased 3D\u2010shape reconstruction from X\u2010ray images. InProc. ICPR(2006) pp.371\u20134. 2","DOI":"10.1109\/ICPR.2006.279"},{"key":"e_1_2_8_24_2","unstructured":"MagnorM. KindlmannG. HansenC.: Constrained inverse volume rendering for planetary nebulae. InProc. VIS(2004) pp.83\u201390. 2"},{"key":"e_1_2_8_25_2","doi-asserted-by":"publisher","DOI":"10.1109\/TBME.2004.827537"},{"key":"e_1_2_8_26_2","doi-asserted-by":"crossref","unstructured":"NarihiraT. MaireM. YuS. X.: Direct intrinsics: Learning albedo\u2010shading decomposition by convolutional regression. InProc. CVPR(2015) pp.2992\u20132992. 3","DOI":"10.1109\/ICCV.2015.342"},{"key":"e_1_2_8_27_2","doi-asserted-by":"crossref","unstructured":"PepikB. BenensonR. RitschelT. SchieleB.: What is holding back convnets for detection? InProc. GCPR(2015) pp.517\u201328. 3","DOI":"10.1007\/978-3-319-24947-6_43"},{"key":"e_1_2_8_28_2","doi-asserted-by":"crossref","unstructured":"QiC. R. SuH. NiessnerM. DaiA. YanM. GuibasL. J.: Volumetric and multi\u2010view cnns for object classification on 3d data. InCVPR(2016) pp.5648\u20135656. 3","DOI":"10.1109\/CVPR.2016.609"},{"key":"e_1_2_8_29_2","unstructured":"RezendeD. J. EslamiS. A. MohamedS. BattagliaP. JaderbergM. HeessN.: Unsupervised learning of 3d structure from images. InNIPS(2016) pp.4996\u20135004. 2"},{"key":"e_1_2_8_30_2","doi-asserted-by":"crossref","unstructured":"RonnebergerO. FischerP. BroxT.: U\u2010net: Convolutional networks for biomedical image segmentation. InInt. Conf. Medical Image Computing and Computer\u2010Assisted Intervention(2015) pp.234\u201341. 4","DOI":"10.1007\/978-3-319-24574-4_28"},{"key":"e_1_2_8_31_2","doi-asserted-by":"crossref","unstructured":"RamamoorthiR. HanrahanP.: A signal\u2010processing framework for inverse rendering. InSIGGRAPH(2001) pp.117\u201328. 2","DOI":"10.1145\/383259.383271"},{"issue":"3","key":"e_1_2_8_32_2","first-page":"1","article-title":"Learning representations by back\u2010propagating errors","volume":"5","author":"Rumelhart D. E.","year":"1988","journal-title":"Cognitive modeling"},{"key":"e_1_2_8_33_2","doi-asserted-by":"crossref","unstructured":"RematasK. RitschelT. FritzM. Gavve sE. TuytelaarsT.: Deep reflectance maps. InProc. CVPR(2016) pp.4508\u201316. 3","DOI":"10.1109\/CVPR.2016.488"},{"key":"e_1_2_8_34_2","doi-asserted-by":"crossref","unstructured":"SerradellE. RomeroA. LetaR. GattaC. Moreno\u2010NoguerF.: Simultaneous correspondence and non\u2010rigid 3d reconstruction of the coronary tree from single x\u2010ray images. InICCV(2011) pp.850\u2013857. 2","DOI":"10.1109\/ICCV.2011.6126325"},{"key":"e_1_2_8_35_2","doi-asserted-by":"publisher","DOI":"10.1109\/TMI.1982.4307558"},{"key":"e_1_2_8_36_2","doi-asserted-by":"crossref","unstructured":"TatarchenkoM. DosovitskiyA. BroxT.: Octree generating networks: Efficient convolutional architectures for high\u2010resolution 3d outputs.arXiv:1703.09438(2017). 2","DOI":"10.1109\/ICCV.2017.230"},{"key":"e_1_2_8_37_2","doi-asserted-by":"publisher","DOI":"10.1109\/TIP.2003.819861"},{"key":"e_1_2_8_38_2","doi-asserted-by":"crossref","unstructured":"W\u00fcrflT. GhesuF. C. ChristleinV. MaierA.: Deep learning computed tomography. InMedical Image Computing and Computer\u2010Assisted Intervention(2016) pp.432\u201340. 3","DOI":"10.1007\/978-3-319-46726-9_50"},{"key":"e_1_2_8_39_2","doi-asserted-by":"crossref","unstructured":"WangW. HuangQ. YouS. YangC. NeumannU.: Shape inpainting using 3d generative adversarial network and recurrent convolutional networks.arXiv:1711.06375(2017). 2","DOI":"10.1109\/ICCV.2017.252"},{"key":"e_1_2_8_40_2","doi-asserted-by":"crossref","unstructured":"WengerS. LorenzD. MagnorM.: Fast image\u2010based modeling of astronomical nebulae.Comp. Graph. Forum (Proc. PG)32 7 (2013) 93\u2013100. 5 6","DOI":"10.1111\/cgf.12216"},{"key":"e_1_2_8_41_2","unstructured":"World Health Organization:Baseline country survey on medical devices 2011. 2"},{"key":"e_1_2_8_42_2","unstructured":"WuZ. SongS. KhoslaA. YuF. ZhangL. TangX. XiaoJ.: 3d shapenets: A deep representation for volumetric shapes. InCVPR(2015) pp.1912\u201320. 2"},{"key":"e_1_2_8_43_2","unstructured":"WuJ. ZhangC. XueT. FreemanB. TenenbaumJ.: Learning a probabilistic latent space of object shapes via 3d generative\u2010adversarial modeling. InNIPS(2016) pp.82\u201390. 2"}],"container-title":["Computer Graphics Forum"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/api.wiley.com\/onlinelibrary\/tdm\/v1\/articles\/10.1111%2Fcgf.13369","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1111\/cgf.13369","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,9,23]],"date-time":"2023-09-23T16:36:50Z","timestamp":1695487010000},"score":1,"resource":{"primary":{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/10.1111\/cgf.13369"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2018,5]]},"references-count":42,"journal-issue":{"issue":"2","published-print":{"date-parts":[[2018,5]]}},"alternative-id":["10.1111\/cgf.13369"],"URL":"https:\/\/doi.org\/10.1111\/cgf.13369","archive":["Portico"],"relation":{},"ISSN":["0167-7055","1467-8659"],"issn-type":[{"value":"0167-7055","type":"print"},{"value":"1467-8659","type":"electronic"}],"subject":[],"published":{"date-parts":[[2018,5]]},"assertion":[{"value":"2018-05-22","order":2,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}