search

Huangying-Zhan / Depth-VO-Feat

Unsupervised Learning of Monocular Depth Estimation and Visual Odometry with Deep Feature Reconstruction
Other
349 stars 66 forks source link

Transform relative location to absolute location in SfMLearner #36

Open HuangGuoHui110 opened 3 years ago

HuangGuoHui110 commented 3 years ago

Hi, as we know, all the predictions in SfMLearner are 5-frame snippets with the format of timestamp tx ty tz qx qy qz qw consistent with the TUM evaluation toolkit. However, the odometry results which you had released are continuous absolute position. How did you convert the relative position to ensure the accuracy of the results? Thanks!

Huangying-Zhan commented 3 years ago

Hi, please see this script for getting the sfmlearner result. https://github.com/Huangying-Zhan/Depth-VO-Feat/blob/master/tools/sfmlearner_odometry_tool/get_sfmlearner_result.py

HuangGuoHui110 commented 3 years ago

Thanks for your reply. : )