# 2D3D registration

**URL:** https://discourse.itk.org/t/2d3d-registration/5533
**Category:** Algorithms
**Tags:** 2d3d, registration
**Created:** [November 30, 2022, 2:02pm UTC](https://discourse.itk.org/t/2d3d-registration/5533 "2022-11-30T14:02:14Z")
**Posts on this page:** 2
**Page:** 1

<div class="post-metadata">

### Author: ![suzume](https://discourse.itk.org/letter_avatar_proxy/v4/letter/s/6a8cbe/32.png) [@suzume](https://discourse.itk.org/u/suzume)
#### Post date: [November 30, 2022, 2:02pm UTC](https://discourse.itk.org/t/2d3d-registration/5533/1 "2022-11-30T14:02:15Z")

</div>

Hi, I want to register the 3D CT image with the 2D image. The IntensityBased2D3DRegistration.cxx needs a pair of 2D orthographic images. And I only have one 2D image. Can I treat this 2D image as a 3D image and register it using Euler3DTransform with a 3D CT image? Or what can I do to register these images and get the six parameters ( translation along x, y , z axes and rotations around x, y , z axes)?

---

<div class="post-metadata">

### Author: ![zivy](https://discourse.itk.org/user_avatar/discourse.itk.org/zivy/32/1726_2.png) [@zivy](https://discourse.itk.org/u/zivy)
#### Post date: [November 30, 2022, 3:01pm UTC](https://discourse.itk.org/t/2d3d-registration/5533/2 "2022-11-30T15:01:38Z")

</div>

Hello @suzume,

The [IntensityBased2D3DRegistration.cxx](https://github.com/InsightSoftwareConsortium/ITKApps/blob/master/IntensityBased2D3DRegistration/IntensityBased2D3DRegistration.cxx) expects images that were created via a [perspective projection](https://en.wikipedia.org/wiki/3D_projection#Perspective_projection), not an [orthographic projection](https://en.wikipedia.org/wiki/3D_projection#Orthographic_projection).

You should not treat the 2D image as a tomographic 3D image, because it isn’t.

Based on the question, I highly recommend that you gain a better understanding of the theory behind 2D/3D (perspective projection image / tomographic image), before working on an implementation. A good place to start is the classical review by Markelj et al., “A review of 3D/2D registration methods for image-guided interventions”, Medical Image Analysis 16(3): 642-661.