how to create rendering_metadata.txt?

[gopikrishnachaganti]

How to generate rendering_metadata.txt for the specific model for the training?

[KnightOfTheMoonlight]

+1

[gopikrishnachaganti]

Can anybody help on how to generate rendering_metadata?

[xiayizhan2017]

hi， I want to know the theory of converting cloud data into two-dimensional images.

[dclcs]

I have tested in the 3D processing software.

Just render a model , and generate the specifice image. I use the camera params and camera model algorithmes to get the image point.

[gopikrishnachaganti]

Good to know that you have already made it, can you elaborate on the steps you did it if possible?

[YiruS]

I have tested in the 3D processing software.

Just render a model , and generate the specifice image. I use the camera params and camera model algorithmes to get the image point.

Same question here. Do you use blender and which kind of camera params are used here? Thanks!

[KnightOfTheMoonlight]

I think they use prior camera knowledge to set the camera parameters. From my understaning, they use azimuth (0-360), elevation around 30, tilt 0, camera distance to the zero of an object around 0.8m.

[walsvid]

Hi everyone. Since the dataset we use is also from the author of 3DR2N2, the original data was rendered using blender. The camera parameters for rendering are just like @KnightOfTheMoonlight said.

[markuspaschi]

I wrote a Master Thesis about 3D-Methods including Pixel2Mesh (in German). But my repository can be found here: https://github.com/markuspaschi/ShapeNetTools Go into DataSet_Tools/Renderer to find hints about how I used the blender renderer to create images. I changed some parameters (no guarantee that it is working better, but it may be a starting point for some people)

[STASYA00]

@KnightOfTheMoonlight @walsvid In the rendering_metadata.txt there are 5 parameters. so, 1 - azimuth 2 - elevation 3 - tilt 4 - distance to the zero to the object 5 - ?

[Michaelwjh]

I think they use prior camera knowledge to set the camera parameters. From my understaning, they use azimuth (0-360), elevation around 30, tilt 0, camera distance to the zero of an object around 0.8m.

But the rendering_metadata.txt there are 5 parameters. What's the last parameter