Single-image Tomography: 3D Volumes from 2D Cranial X-Rays
| dc.contributor.author | Henzler, Philipp | en_US |
| dc.contributor.author | Rasche, Volker | en_US |
| dc.contributor.author | Ropinski, Timo | en_US |
| dc.contributor.author | Ritschel, Tobias | en_US |
| dc.contributor.editor | Gutierrez, Diego and Sheffer, Alla | en_US |
| dc.date.accessioned | 2018-04-14T18:24:57Z | |
| dc.date.available | 2018-04-14T18:24:57Z | |
| dc.date.issued | 2018 | |
| dc.description.abstract | As many different 3D volumes could produce the same 2D x-ray image, inverting this process is challenging. We show that recent deep learning-based 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-resolution volume which is then fused in a second step with the input x-ray into a high-resolution volume. To train and validate our approach we introduce a new dataset that comprises of close to half a million computer-simulated 2D x-ray images of 3D volumes scanned from 175 mammalian species. Future applications of our approach include stereoscopic rendering of legacy x-ray images, re-rendering of x-rays 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-rays. | en_US |
| dc.description.number | 2 | |
| dc.description.sectionheaders | Image Magic | |
| dc.description.seriesinformation | Computer Graphics Forum | |
| dc.description.volume | 37 | |
| dc.identifier.doi | 10.1111/cgf.13369 | |
| dc.identifier.issn | 1467-8659 | |
| dc.identifier.pages | 377-388 | |
| dc.identifier.uri | http://dx.doi.org/10.1111/cgf.13369 | |
| dc.identifier.uri | https://diglib.eg.org:443/handle/10.1111/cgf13369 | |
| dc.publisher | The Eurographics Association and John Wiley & Sons Ltd. | en_US |
| dc.subject | Deep learning | |
| dc.subject | Volume rendering | |
| dc.subject | Inverse rendering | |
| dc.subject | Convolutional neural networks | |
| dc.subject | Tomography | |
| dc.title | Single-image Tomography: 3D Volumes from 2D Cranial X-Rays | en_US |