Data transforms.py

[hyunJIN7]

Hi, I am a big fan of this project. So I'm trying to study NeuralRecon code.

I have some questions about tranforms.py

First, What is the self.random_rotation and self.random_translation becoms False? Why do you apply a random 3x4 linear transform to the world coordinate system without using the default value( r=0, t=.5 )? Would you explain how this specifically affect pose and TSDFs?

Second, what is the meaning of randomly sample a crop? I don't understand from line 209 to line 215. Could you explain this in more detail?

Thanks, Hyunjin Kim.

[MingyunKang98]

안녕하세요~ 이쪽 CV분야에 관심있는 학생입니다. neural recon 관련하여 조금 자세한 질문을 드리고 싶습니다. e-mail로 여쭤보고 싶은데 주소를 받을 수 있을까요?

[Rudramani4493]

Hi, I am a big fan of this project. So I'm trying to study NeuralRecon code.

I have some questions about tranforms.py

First, What is the self.random_rotation and self.random_translation becoms False? Why do you apply a random 3x4 linear transform to the world coordinate system without using the default value( r=0, t=.5 )? Would you explain how this specifically affect pose and TSDFs?

Second, what is the meaning of randomly sample a crop? I don't understand from line 209 to line 215. Could you explain this in more detail?

Thanks, Hyunjin Kim.

Hi, I am a big fan of this project. So I'm trying to study NeuralRecon code.

I have some questions about tranforms.py

First, What is the self.random_rotation and self.random_translation becoms False? Why do you apply a random 3x4 linear transform to the world coordinate system without using the default value( r=0, t=.5 )? Would you explain how this specifically affect pose and TSDFs?

Second, what is the meaning of randomly sample a crop? I don't understand from line 209 to line 215. Could you explain this in more detail?

Thanks, Hyunjin Kim.

Hi Hyunjin,

Did you got the answer about how the linear transformation effects the pose and tsdfs?