Question

I am reading a lot about 3D reconstruction, as I want to implement it for my product using OpenCV. A thing I want to confirm is if it's possible to do 3D reconstruction of a changing scene. I will explain better.

All of the articles talk about a moving camera or multiple cameras taking shots of the same scene. All of these shots are then converted to meshes that are merged into one by using features as connection points.

Is it possible to do the same with a fixed camera and rotating subject? I think features will still be working, right? The only thing I'm not sure is if the changing background will be a problem. Will I have to recognize and extract the object I want to reconstruct to eliminate noise maybe?

Thanks

Was it helpful?

Solution

Is it possible to do the same with a fixed camera and rotating subject?

Yes. Moving camera or moving scene is essentially the same thing. It's just a matter of reference.

The only thing I'm not sure is if the changing background will be a problem. Will I have to recognize and extract the object I want to reconstruct to eliminate noise maybe?

Yes, a non-static scene is more problematic. You can solve this by detecting which parts are static, and which are dynamic, and then only use the static objects. It is however not trivial.

Licensed under: CC-BY-SA with attribution
Not affiliated with StackOverflow
scroll top