leggedrobotics / wild_visual_navigation

Wild Visual Navigation: A system for fast traversability learning via pre-trained models and online self-supervision
https://sites.google.com/leggedrobotics.com/wild-visual-navigation
MIT License
113 stars 9 forks source link

When undistorting please provide new camera matrix to store it with the images. #14

Closed JonasFrey96 closed 2 years ago

JonasFrey96 commented 2 years ago

@mmattamala

mmattamala commented 2 years ago

Ok, I can check that tomorrow with the real calibration files of the cameras (if we have them somewhere). Worst case we calibrate those alphasenses again.

mmattamala commented 2 years ago

@JonasFrey96 I added this to the latest commit, you can check the example python file to check the usage. I'll close the issue

JonasFrey96 commented 2 years ago

@mmattamala fully integrated and tested the feature in: https://github.com/leggedrobotics/anymal_dataset_generation/tree/feature/drs_data_tools Thanks a lot.