search

zhangcv123 / MBRVO-Dataset

14 stars 0 forks source link

Format of Datasets #1

Closed imuncle closed 25 minutes ago

imuncle commented 2 hours ago

Thanks for the great work! I have downloaded the datasets and have some questions.

What's the unit of depth map (mm?)

What's the format of camera pose (is the first three numbers stand for translation and later four numbers for quaternion)? If so, what is the unit of translation, and how the frame of camera is defined? dose the x axis towards right, y axis towards down, z axis towards front?

Please forgive me for asking so many questions. Hope you can give me answers.

By the way, I didn't find optical flow ground truth in datasets :(

zhangcv123 commented 2 hours ago

Thanks for the great work! I have downloaded the datasets and have some questions.

What's the unit of depth map (mm?)

What's the format of camera pose (is the first three numbers stand for translation and later four numbers for quaternion)? If so, what is the unit of translation, and how the frame of camera is defined? dose the x axis towards right, y axis towards down, z axis towards front?

Please forgive me for asking so many questions. Hope you can give me answers.

By the way, I didn't find optical flow ground truth in datasets :(

Thank you for following our work. The unit of the depth map is m, the camera pose is x y z qx qy qz qw, and the camera coordinate system is UE. After a while, I will upload the entire dataset and update detailed instructions in the README.

imuncle commented 1 hour ago

Thanks a lot!