# 3D Reconstruction from 2D XRay DICOM images

**URL:** <https://discourse.itk.org/t/3d-reconstruction-from-2d-xray-dicom-images/7161>\
**Category:** Beginner Questions\
**Tags:** dicom\
**Created:** [August 8, 2024, 10:18am UTC](https://discourse.itk.org/t/3d-reconstruction-from-2d-xray-dicom-images/7161 "2024-08-08T10:18:44Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![drathin4](https://discourse.itk.org/user_avatar/discourse.itk.org/drathin4/32/4333_2.png) [@drathin4](https://discourse.itk.org/u/drathin4)\
**Post date:** [August 8, 2024, 10:18am UTC](https://discourse.itk.org/t/3d-reconstruction-from-2d-xray-dicom-images/7161/1 "2024-08-08T10:18:44Z")

</div>

Can we reconstruct 3D model from 2D DiCoM images obtained from XRays?

I read this link, but didn’t find any relevant answer:  
[3D reconstruction](https://discourse.itk.org/t/how-can-i-reconstruct-series-or-sequence-of-2d-dicom-images-into-3d-reconstruction/3495)

I also didn’t find any geometric metadata in the XRay DiCoM files unlike that available in CT/MRI files.

---

<div class="post-metadata">

**Author:** ![dzenanz](https://discourse.itk.org/user_avatar/discourse.itk.org/dzenanz/32/1093_2.png) [@dzenanz](https://discourse.itk.org/u/dzenanz)\
**Post date:** [August 8, 2024, 12:26pm UTC](https://discourse.itk.org/t/3d-reconstruction-from-2d-xray-dicom-images/7161/2 "2024-08-08T12:26:48Z")

</div>

That is a non-trivial task, take a look at one example of it being done: [https://hal.science/hal-02181752/file/IBHGC\_ESJO\_2019\_GAJNY.pdf](https://hal.science/hal-02181752/file/IBHGC_ESJO_2019_GAJNY.pdf)

---

<div class="post-metadata">

**Author:** ![drathin4](https://discourse.itk.org/user_avatar/discourse.itk.org/drathin4/32/4333_2.png) [@drathin4](https://discourse.itk.org/u/drathin4)\
**Post date:** [August 8, 2024, 2:13pm UTC](https://discourse.itk.org/t/3d-reconstruction-from-2d-xray-dicom-images/7161/3 "2024-08-08T14:13:35Z")

</div>

I have read similar research papers which used techniques of statistical shape modeling (SSM) and contour extraction from freeform deformation (FFD). But, that seems out of the scope.

I was wondering if getting depth information for 3D reconstruction is possible from camera parameters (which are not available at the moment) and then using image processing techniques.