Why some images’ kptsmat only has 10 dimensions instead of 11？How to reconstruct the 3D keypoints？

JotaBravo / spacecraft-uda

Spacecraft Pose Estimation: Robust 2D and 3D-Structural Losses and Unsupervised Domain Adaptation by Inter-Model Consensus

https://ieeexplore.ieee.org/document/10225381

MIT License

19 stars 1 forks source link

Why some images’ kptsmat only has 10 dimensions instead of 11？How to reconstruct the 3D keypoints？ #12

Closed LiKrP closed 8 months ago

LiKrP commented 8 months ago

Hi，thanks for your nice work. The detailed preprocessed data saved me a lot of time.

I'm confused about two things. I notice that some images’ kptsmat only has 10 dimensions instead of 11, does this has some special meaning？ I also notice that the groundtruth of the 3D kpts are very “neat”, what method did you use to reconstruct?

Thank you.

JotaBravo commented 8 months ago

Hi,

Can the missing keypoints occur because they fall outside the image and thus not included? I'm afraid I was not aware of this issue. I'll have a look more in detail.

Regarding the 3D keypoints. We did something similar as the process described here:

https://openaccess.thecvf.com/content_ICCVW_2019/html/R6D/Chen_Satellite_Pose_Estimation_with_Deep_Landmark_Regression_and_Nonlinear_Pose_ICCVW_2019_paper.html