The generation of test1_files.txt for the KITTI dataset

[jibowang26]

Dear Prof. Shi,

Thank you so much for sharing the code of your paper. Your paper is very interesting and gives me great inspiration.

When I tested the KITTI dataset using this command python train_kitti_3DoF.py --batch_size 1 --test 1 , the program needs to load two files : dataLoader/test1_files.txt, dataLoader/test2_files.txt . The data in these two files includes the filename, gt_sthift_x, gt_sthift_y, and theta. I would like to inquire about how gt_sthift_x, gt_sthift_y, and theta were generated and how I should generate the test1_files.txt or test2_files.txt? Could you provide a method or program to generate the data files test1_files. txt or test2_files.txt?

Thank you and looking forward to your reply.

Best regards,
Jibo Wang

[dizzyrue]

same question

[YujiaoShi]

plz refer to this link: https://github.com/YujiaoShi/HighlyAccurate.git

[jibowang26]

plz refer to this link: https://github.com/YujiaoShi/HighlyAccurate.git

Thank you very much for your reply, Best wishes.