You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
二、论文中指出(包括OnePose++)only a video sequence with annotated object poses is available。请问此处的annotated object pose是指什么?算法不是只使用了video sequence 吗,图像的位姿不是可以在SFM过程求出吗
期待您的回复,
非常感谢!
The text was updated successfully, but these errors were encountered:
感谢您的工作!
在拜读您论文时,有两个问题不太明白,特向您请教:
一、请问使用AR工具标注Bounding Box的作用是什么?
我是这样理解的:
1. SFM重建的点云是相机坐标系下的,使用BBox将其转换到目标坐标系下?
2. 多次采集的图像序列重建的点云有偏差,使用BBox将它们融合?
如果第1个理解正确的话,那么使用无AR功能的相机是不是OnePose就不能正常工作了?
二、论文中指出(包括OnePose++)only a video sequence with annotated object poses is available。请问此处的annotated object pose是指什么?算法不是只使用了video sequence 吗,图像的位姿不是可以在SFM过程求出吗
期待您的回复,
非常感谢!
The text was updated successfully, but these errors were encountered: