Why set temporal frames from default 2 to 10 when 'lidar_depth' and 'annotated_lidar' are enabled?

[Rui Li]

Hi,

Thank you al lot for your impressive work.

I found a place that I do not understand. As shown in the code below

MonoRec/data_loader/kitti_odometry_dataset.py

Lines 56 to 58 in b2b3dec

 if self.annotated_lidar and self.lidar_depth: 
     extra_frames = max(extra_frames, 10) 
     self._offset = max(self._offset, 5) 

why set the nearby frames from default 2 to at least 10 when 'lidar_depth' and 'annotated_lidar' are enabled. Is it a general setting for both training and testing or only a setting for testing? Thank you!

[Felix Wimbauer]

Thank you for showing interest in our work!

If I remember correctly, this is because the annotated lidar is not available for the 5 first and 5 last frames of a sequence. (they are generated by aggregating lidars over multiple timesteps).
Therefore, we have to make sure to ignore the first and last 5 frames in the sequence (-> extra frames >= 10, offset >= 5).

Best,
Felix

[Rui Li]

Oh I understand, so the number of adjacent frames is not changed (==2), and only the first and last 5 frames are overlooked. Do I understand it correctly? Best

…

On Wed, Jul 27, 2022 at 9:40 PM Felix Wimbauer ***@***.***> wrote: Thank you for showing interest in our work! If I remember correctly, this is because the annotated lidar is not available for the 5 first and 5 last frames of a sequence. (they are generated by aggregating lidars over multiple timesteps). Therefore, we have to make sure to ignore the first and last 5 frames in the sequence (-> extra frames >= 10, offset >= 5). Best, Felix — Reply to this email directly, view it on GitHub <#40 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AFL6463PE6SBASKTHWRBXE3VWE35DANCNFSM54ZUNXOQ> . You are receiving this because you authored the thread.Message ID: ***@***.***>

[Felix Wimbauer]

You are right :)