Volumetric Capture with RealSense Depth Cameras!

Recently at the lab we got some new Intel RealSense Depth Cameras. These cameras are able to create point clouds of real-world scenes. In other words, the data produced consists of 3-dimensional points in space! There are many possible use cases for this, but one of the ones we have been focusing on here at the lab is the possibility of having a lecturer be remotely broadcasted to students.

With the functionality of the Depth Cameras, we can easily track and make a 3D recreation of people and objects in the scene as we can see below:

2D Depth Camera compared with an RGB Camera
3D Recreation with Texture Applied

3D Recreation with Depth Texture Applied

Further engineering will allow us to create 3D meshes from these point clouds, if necessary. Additionally, with multiple cameras we will be able to use algorithms to minimize the error when syncing point clouds together.

Hope you enjoyed the little sneak peek at what we are doing, and hope to write again in two weeks! : )

Leave a Comment

Your email address will not be published.