I have generated dataset with nvidia isaac sim from custom fbx models. But the major issue is, I am getting images with occlusion. Hence the detection quality is decreasing when two or more packages are stacked together. I would like to know how to avoid that occlusion while I am generating the dataset.
Hi @sk.ahmed401
To avoid occlusion while generating dataset, you will have to add heuristics for spawning of assets given a camera viewpoint