Your depth map formula requires the baseline. So, how can you obtain depth map for mono-images (which are obtained without stereo cameras)

[sarimmehdi]

Hello. I noticed in your research paper, the formula you use to get the z value (depth value) requires the baseline. But you also claim you are able to obtain point clouds for mono depth maps. But, in that case, there is no baseline. Then how are you able to do that?

[mileyan]

Hi @sarimmehdi , the stereo models will predict disparities, instead of depth. We need the baseline to project disparity to depth. However, the monocular models can directly predict depth, so you don't need the baseline anymore.