Generate mesh from depth images

Hi, I’m not as familiar as i would like to be with Isaac Sim. I was wondering if Isaac Sim has functionality built in to A, generate mesh at runtime and B, is there already code to generate mesh at runtime using a depth image? The idea is to construct the scene as we move around with our camera.
I tried having a look around but haven’t seen anything specifically about doing this during runtime/simulation.


Hi there,

I am not aware of any example doing this automatically, here is however an example how you can construct a mesh in USD once you have the vertex details (e.g. point clouds from depth data). You can also use semantic segmentation of the images to split the point clouds into relevant parts and then build the mesh.

You can also take a look at nvblox on how you could do this: