Skip to content

Workflow Stage 1: Capturing

by on February 28, 2012

Capturing is the process of grabbing data from Kinect sensor to be used by Matherix 3Dify to build  3D model. Check out this tutorial video for more info.

Some of the good practices while capturing video are:

1> Segment the object of interest from the background using the slider.
At the start of the capturing process, you are given a slider to adjust the depth. By adjusting it, you can segment away the unnecessary background(shown in white) . Try and keep only the object of interest in the scene as shown below.

2> Avoid flat regions like a wall or floor.
It is always preferable to avoid flat regions like a wall or floor. In case, you cannot avoid it for some reason like a flower pot placed on the floor, a simple trick is to place some crumpled cloth around the object. The crumpled-ness creates more depth discontinuities (or) bumps which are actually beneficial in the Alignment stage. We can remove these crumpled-cloth sections from our final model at the end in the post processing stage. A rule of thumb is that more the non-flatness of the scene, better would the result be.

3> Avoid direct light at the Kinect.

Always try and keep any form of lights (natural or otherwise) behind your Kinect and not in front of it. Direct light at Kinect introduces lot of noise which can hinder in alignment and surely require more work at Noise-removal stage. It is a good practice to keep a good ambient light with no direct light at Kinect cameras. So, avoid  scenarios shown below:

We will shortly put up tutorials for good practices for various types of objects like a face, shoe, furniture, plant etc.

~Matherix team


From → Tutorial

Leave a Comment

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: