search

HpWang-whu / RoReg

[TPAMI 2023] RoReg: Pairwise Point Cloud Registration with Oriented Descriptors and Local Rotations
https://hpwang-whu.github.io/RoReg/
90 stars 12 forks source link

Preprocessed dataset #2

Closed zhaoyiww closed 1 year ago

zhaoyiww commented 1 year ago

Hi, thanks for the amazing work! I have questions about your preprocessed 3DMatch datasets. Are the keypoints generated from the work FCGF? And I notice you mentioned keypoint detection in the Method section of the paper, so I also wonder how these keypoints (from the preprocessed datasets) are used in your algorithm. Best regards

HpWang-whu commented 1 year ago

Hi @zhaoyiww， Thanks for your interest in our work! Yes, we use the same preprocessed data as FCGF\SpinNet\YOHO. We first extract RoReg-Desc on the randomly sampled ‘’keypoints‘’ in the preprocessed data (5k in each scan). Then, we conduct keypoint selection utilizing these RoReg-Desc.

Yours,

zhaoyiww commented 1 year ago

Thanks for the reply!