Can anyone suggest an algorithm to build a 3D image from multiple 2D images (regular pics, from mobile robots cameras). Algorithm itself or information about .
If emailing, remove "SPAM NO MORE" in my addresses (repeated twice). THANKS IN ADVANCE, THIS IS A COLLEGE PROJECT, WILL INCLUDE YOU ON THE LIST OF REFERNCES
Try the extended polygon method. Take image number one, create its outline, and extend it from an imaginary perspective point behind the robot's "eye". This creates an irregular prism that radiates away from the eye to the distance. Now take the second view and do the same. It will create a second radiating prism that intersects the first. This trims the solid to a closer approximation of the true object. Repeat with each view. After three to five views, you end up with a clipped polytope that is a fair representation of the object. And, if you have depth information from (for example) radar or ultrasound, this can become the from clip limit of the polytope.
Cheers!
Sir Charles W. Shults III, K. B. B. Xenotech Research
PolyTech Forum website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.