ShuweiShao / AF-SfMLearner

[MedIA2022 & ICRA2021] Self-Supervised Monocular Depth and Ego-Motion Estimation in Endoscopy: Appearance Flow to the Rescue
MIT License
99 stars 15 forks source link

SCARED point cloud acc and EndoSLAM #45

Closed shanwq closed 8 months ago

shanwq commented 1 year ago

作者您好! 我想问一下,scared数据集的点云精度是如何得到的呀?将depth转为点云之后如何计算accuracy呢? 还有就是EndoSLAM的五折交叉验证,是用上了LowCam所有的帧吗?因为我看您似乎没有用scared所有的帧。 以上两部分的结果,不知道您方不方便提供一下代码。如果可以的话,真是非常感谢了。如有打扰,还望海涵!

ShuweiShao commented 11 months ago

您好,不好意思,刚注意到这个问题。Accuracy这儿指的是RMSE,与深度误差的计算类似。EndoSLAM的五折交叉验证用了所有的帧。代码我找找看,时间有点长了。