Dear @IIPCV-LAB:
Thanks for sharing your codes. Your method is amazing. However, do you have implementation about your compared method (such as CalibNet)? Actually the author of CalibNet used a different dataset (KITTI raw dataset 2011_09_26_09) with yours (KITTI Odometry). So I am really interested about how your compared methods performs under your experimental environment.
Hope for your reply, thank you!
Cheers!