I'm a lifelong member of Polycam. Could you make a specific lecture video on how to trim the material? I also want to know which method you usually use.
this is a cool and easy way to use a photo scan to create a character but it's not a good practice if you wanna make a game. The topology is a mess in the photoscans which makes the mesh heavy, the textures appear bad and the stretching of the mesh inconsistent. You can use your photo scans to create a metahuman face mesh. That way you'll get a fully rigged, textured character with good topology. Nanite does not (yet) work with skeletal meshes.
nanite is for static mesh (mesh with no skeleton) atm so will not work with moving objects even if it does work with foliage in 5.1 wich is already a revolution.
low poly version is bcz UE = game engine so we use tricks. Idea here is to project the textures in HQ (nomal map etc) on a low poly version. Meaning you see the same level of details but its optimized for real time and also for movements, bad topology fuck up the mesh as my collegue explained before me
I love your app but this tutorial is full of BAD ADVICES if not just NONSENSE. Sorry for CAPS. For the bad advices : 1. You can't have a decent result with photogrammetry and full human because it's too slow --> subject will move --> scan will be messy. --> Instead use video (RAW format appréciated if you can), and for the --> face only
A top knot AND saying "how meta is that" in one video, fantastic, I feel like it's my birthday
What a detailed tutorial bro, congrats!
Fantastic, thanks for the tut
I love how the models of the characters always end up looking like those roblox meshes
nice video, thanks
For some reason after exporting mine is not with mesh when I upload to the mixamo website any ideas?
me too
I'm a lifelong member of Polycam.
Could you make a specific lecture video on how to trim the material?
I also want to know which method you usually use.
Check out Blender tutorials on that. 1000’s of videos on what you’re asking for
I assume you don't have to create a low poly version if you're using Nanite right?
this is a cool and easy way to use a photo scan to create a character but it's not a good practice if you wanna make a game. The topology is a mess in the photoscans which makes the mesh heavy, the textures appear bad and the stretching of the mesh inconsistent. You can use your photo scans to create a metahuman face mesh. That way you'll get a fully rigged, textured character with good topology. Nanite does not (yet) work with skeletal meshes.
nanite is for static mesh (mesh with no skeleton) atm so will not work with moving objects even if it does work with foliage in 5.1 wich is already a revolution.
low poly version is bcz UE = game engine so we use tricks.
Idea here is to project the textures in HQ (nomal map etc) on a low poly version.
Meaning you see the same level of details but its optimized for real time and also for movements, bad topology fuck up the mesh as my collegue explained before me
Awesome, I am so satisfied
Kinda awkward considering Epic is making its own (better) photogrammetry app.
Whats it called?
I love your app but this tutorial is full of BAD ADVICES if not just NONSENSE.
Sorry for CAPS.
For the bad advices :
1. You can't have a decent result with photogrammetry and full human because it's too slow --> subject will move --> scan will be messy.
--> Instead use video (RAW format appréciated if you can), and for the --> face only