I am currently working with the v1.0-mini dataset to test my sample model. Initially, I used the 3D coordinates of LiDAR points as the input for my model. However, I noticed a discrepancy: the maximum x and y values of the LiDAR points are around 100, while the maximum x and y values of the corners in the ground truth boxes are around 600.
Could you please clarify whether these data are in different coordinate systems? If so, how can I transform the data to ensure they are in the same coordinate system?
Has anyone encountered and resolved a similar issue? I am also concerned that I might have misunderstood the data from the Nuscenes benchmark, leading to this problem.
I am working on this for my Master’s thesis, so resolving this issue is very important to me. Thank you in advance for your assistance.