Hi, I am trying to determine what conversion to use to transform the depth matrix into distance (meters from the camera axis to the respective x,y pixel coordinate).
How would you recommend I go about the conversion to actual distance given the depth map?
Example: on the sample image min_depth is 0, max_depth is ~9.06, median_depth is 1.67. Some far away object in the image (probably 10m in real life) has a depth of 1.86.
In other papers they just normalize by using the max observed distance in the training data; I have no idea what that would be in your training data set or if that would be correct to do in this case.
Thank you!
Hi, I am trying to determine what conversion to use to transform the depth matrix into distance (meters from the camera axis to the respective x,y pixel coordinate). How would you recommend I go about the conversion to actual distance given the depth map? Example: on the sample image min_depth is 0, max_depth is ~9.06, median_depth is 1.67. Some far away object in the image (probably 10m in real life) has a depth of 1.86.
In other papers they just normalize by using the max observed distance in the training data; I have no idea what that would be in your training data set or if that would be correct to do in this case. Thank you!