csv mesh

[AOBA200941]

i don't know if you can help me to import csv to blender i found a script but it doesn't impose uvmap in a correct way. 

csv.zip import_pix.py

[AOBA200941]

[shak-otay]

There's lots of doubles, and even worse, many 1.0 1.0 in the vt lines:

...

vt 0.498040 0.498040
vt 0.498040 0.498040
vt 0.498040 0.498040
vt 0.498040 0.498040
vt 0.250980 0.250980
vt 1.000000 1.000000
vt 1.000000 1.000000
vt 1.000000 1.000000
vt 1.000000 1.000000
vt 1.000000 1.000000
vt 1.000000 1.000000
vt 1.000000 1.000000
vt 1.000000 1.000000
vt 1.000000 1.000000
vt 1.000000 1.000000
vt 1.000000 1.000000
vt 1.000000 1.000000
vt 0.250980 0.250980
vt 0.498040 0.498040
vt 0.498040 0.498040
vt 0.498040 0.498040
vt 0.498040 0.498040
vt 0.749020 0.749020
vt 1.000000 1.000000
vt 1.000000 1.000000
vt 1.000000 1.000000
vt 1.000000 1.000000
vt 1.000000 1.000000
vt 1.000000 1.000000

 

Thinking...

Just a wild guess, but I think the uv data interpretation is wrong for these models.

Maybe it's half floats instead of floats or, more likely, a 1-x problem because one half of the uv map is empty. 

edit: reminds me of a problem I'd solved on former xentax; but don't remember, how:

 

Edited March 13 by shak-otay

[AOBA200941]

7 hours ago, shak-otay said:

There's lots of doubles, and even worse, many 1.0 1.0 in the vt lines:

...

vt 0.498040 0.498040
vt 0.498040 0.498040
vt 0.498040 0.498040
vt 0.498040 0.498040
vt 0.250980 0.250980
vt 1.000000 1.000000
vt 1.000000 1.000000
vt 1.000000 1.000000
vt 1.000000 1.000000
vt 1.000000 1.000000
vt 1.000000 1.000000
vt 1.000000 1.000000
vt 1.000000 1.000000
vt 1.000000 1.000000
vt 1.000000 1.000000
vt 1.000000 1.000000
vt 1.000000 1.000000
vt 0.250980 0.250980
vt 0.498040 0.498040
vt 0.498040 0.498040
vt 0.498040 0.498040
vt 0.498040 0.498040
vt 0.749020 0.749020
vt 1.000000 1.000000
vt 1.000000 1.000000
vt 1.000000 1.000000
vt 1.000000 1.000000
vt 1.000000 1.000000
vt 1.000000 1.000000

 

Thinking...

Just a wild guess, but I think the uv data interpretation is wrong for these models.

Maybe it's half floats instead of floats or, more likely, a 1-x problem because one half of the uv map is empty. 

edit: reminds me of a problem I'd solved on former xentax; but don't remember, how:

 

this is a renderdoc capture, maybe they are incomplete because export vs input not vs output was my mistake.

out.zip

[shak-otay]

Well, I should have paid more attention to the first line of the csv!

(simply adjust the indices here:             uv = (float(row[16]), 1.0 - float(row[17]))

 It's for the body.csv from your opening post.)

Edited March 13 by shak-otay

[AOBA200941]

24 minutes ago, shak-otay said:

Well, I should have paid more attention to the first line of the csv!

(simply adjust the indices here:             uv = (float(row[16]), 1.0 - float(row[17]))

 It's for the body.csv from your opening post.)

 

if it works thank you very much

[Karpati]

I released the 3D Object Converter v10.603 (Windows) that supports the following file format:

RenderDoc *.csv for load.

 

How to get the 3D Object Converter v10.603:

- Download the 3D Object Converter from http://3doc.i3dconverter.com and install it or download and use the portable version (if you don’t have it yet).

- Just use the Help/Check for updates... function to get the v10.603.

 

Edited April 9 by Karpati

[Durik256]

On 3/12/2024 at 4:50 AM, AOBA200941 said:

import csv

i made plugin for Noesis fmt_cvs.py (faces, vert, norm, uvs)

[tofulemon]

On 3/12/2024 at 1:05 PM, AOBA200941 said:

ive came across a similar issue, execept i cleared up doubles, and the uv map still looks weird after unwrapping it (the texture for the model is set up differently from the uv map), how can this be solved? (image attached below)