About laser scan reference

Development Asset Creation
,
1

Hi friends, 

i have some component

1 of photogrammetry via Sony

1 of drone

1 of laser scanner

 

i would like to merge the component but i want a laser scanner component like a reference…

 

thanks

 

2

Hi, then you can set four your images Prior pose as Unknown and just keep the position info for laser scan data.

3

Thanks a lot

4

And for laser scanner data i have to define:

locked or position and orientation?

5

With locked the position won’t be changed

6

ok…thanks…

 

for the mesh and texture information, i have to open another post?

7

Don’t have to open another post, you can continue here.

8

ok… thanks

 

for a good mesh qualitie, how can i do?

i have a flat surface a building facade, and the mesh have many edge is not linear…

and the worflow that i do is this…

  • allign (more sensor)

-normal detail

-simplify

-unwrap

-texture…

 

but i have a problem with a regularity of a mesh and a texture resolution

 

understand? my english is bad

9

Can you post here some print screen of your results?

10

why doing thi?

■■■■…

 

11

i have the autosave and the folder of project…why doing this?

12

Just to warn you, problems like this usually get fixed by re-processing the whole project. It’s possible the project file didn’t save correctly and corrupted. But your inputs should still be licensed, It depends if you saved a ppi-licenses.rclicense file when you licensed the images. If so you just need to add this file with the images when you import them. Alternatively the images themselves could hold the license, in which case you just put the images into a new project, and they will be licensed. Either way when the inputs are licensed you should see this L next to them.

There is one solution that may help which is not recommended for other problems, but you can try it now. I recommend making a copy of your project and trying it on that copy. You can open the project file in a text editor (right click on the project file and open in the text editor) and manually delete the last model, and also delete the corresponding .dat file in the project folder. 

13

Ok thanks, i have re - elaborate the project…

 

 

14

i have anotehr question… im view that the model have a not good allineation… i have insert more control point but in one section the alignemnet is not good…

 

it is possible to refine allineament in automatic?

15

After inserting new control points, did you realigned the component?

There could be various issues causing this. Is there enough images in that area? What was the overlap? What were the conditions?

16

now i have put more control point… i attend the mesh…

i have some question, because is the 2 project thath i do with reality capture…

can i also share with you my project? (i haave to put the project file and the folder of project or also the photo)?

  • the question are:
  1. how i check the quality of the allign ? only with a numerical error? i have to do a slice?

2)difference between allign and merge

3)i would like to use a scale of pointcloud and a geometry of point cloud, i have to insert a locked position, right, but in input or in the laser scanner component, before the final alignment?

  1. after  the allignment i would disable a photo for the mesh and texture, i have to disable in the component or in imput?

  2. can i modify a mesh parameters, in this case the normal reconstruction is 680 milion of triangles, i would a less triangle

  3. i create my model, is a god way this workflow

5.1) normal mesh

5.2) simplify mesh

5.3)unwrap

5.4)texture

in this case i don’t have to wait the texture calculate on the normal model or is a incorrect workflow?

 

thanks a lot…

 

 

 

17

Hi, ho wbig is your project, which you want to share? The best way is share everything.

  1. You can use Inspect quality tool for that

https://support.capturingreality.com/hc/en-us/community/posts/4409723827986-Inspection-Tool

2)  Image alignment or image registration is a process in which RealityCapture calculates camera positions, orientations and internal camera states, like camera zooming, for every input image. Once images are registered, you will be able to recognize, how you moved with your camera at the time when you were taking the input images. Use Merge Components tool - merge already created components. When using this function, no new images are added to the existing components

  1. You don’t need to set scale for laser scans, they are already scaled. If you are using just images, then you need to apply the scale before final alignment. Or after placing the scale, you can use Update tool to apply the scale.

  2. It doesn’t matter, when you disable the images for further processing, they will be disabled for the whole project

  3. Yes you can in the Reconstruction settings. But there is no option for a whole maximal triangle count. You can achieve this using Simplify tool

  4. It depends, what you want to achieve. In mostly cases, the highest model is textured, then simplified to a wanted model count and then the texture is reprojected to that smaller model

18

it is indifferente where i disable 1 or 2 the camera for meshing or for texture?

 

and after i have this problem

 

can i share the project?

the link is

https://1drv.ms/u/s!AqWlY7FMvK9yjrtXtLxTvk9ZjzDgNA?e=NhtrEz

19

Yes, it doesn’t matter where you will disable it. It is then disabled for a whole project.

The second issue is bad alignment. You should use some control points in that area to achieve better alignment

I am downloading your project now.

20

yes i had used more control point…