Comments (6)
it's a clipping problem after the training. simple clamping like the code below may circumvent the problem where values spike above 1.0f
def optim_clr(case, img_path, mesh_dir, save_dir, device):
...
np_colors= to_numpy(color_model.colors)
np.clip(np_colors, 0.0, 1.0, out=np_colors) #clamp
tensor_colors = torch.tensor(np_colors ,dtype=torch.float32,device=device)
save_obj(vert, face, f'{save_dir}/refine_{case}.obj', tensor_colors.detach())
...
from era3d.
dunno TBH. would like to take a look into it though. I've been noticing it too. it's weird. with normals flipped in the wrong axes and stuff
from era3d.
@bananaman1983 Excellent! you also solved the issue of noise in the colors of my exported mesh.
by:
texture_color = torch.clamp(color_model.colors.detach(), min=0, max=1)
save_obj(vert, face, f'{save_dir}/refine_{case}.obj', texture_color)
from era3d.
Usually blender can't get vertex color from obj file. ( In my case blender 3.6.2 )
You can export as ply with vertex color from original obj file by tool like MeshLab.
There is also issue about this. #14 (comment)
from era3d.
@bananaman1983 Could you guess about this issue too ? #26
from era3d.
@bananaman1983 @miaowu99 Great Job !
You solved this issue ! #4 👍
before
after
@pengHTYX You can check it too !
from era3d.
Related Issues (20)
- I want to adjust Advanced Options HOT 1
- Question about Cross-domain Att HOT 4
- The problem of the mesh being mirrored left and right HOT 1
- Instant-NSR Mesh Extraction 失败,我是windows平台 HOT 5
- Can I run the Python app. py on Windows? HOT 2
- This is a very good work. May I ask what the camera intrinsic parameter matrix and fovy are corresponding to the six images generated? HOT 2
- I wonder if we can generate 3D models HOT 2
- The shape of the character in the image is not correctly captured. HOT 1
- Can we use free viewpoint? HOT 1
- WARNING: Requirement 'xformers-0.0.23.post1-cp39-cp39-manylinux2014_x86_64.whl' looks like a filename, but the file does not exist ERROR: xformers-0.0.23.post1-cp39-cp39-manylinux2014_x86_64.whl is not a supported wheel on this platform. HOT 1
- IndexError: The shape of the mask [1, 64, 64] at index 1 does not match the shape of the indexed tensor [1, 360, 360, 3] at index 1
- Can I create additional images from another angles ? HOT 9
- position embeddings inconsistency HOT 1
- Why are the resulting objects flatter, less round, and have white patches on their sides?
- Can you provide the training code as Wonder3D? HOT 2
- Commercial use? HOT 2
- What is the unit of estimated elevation? HOT 1
- Time complexity for inference
- Question about number of views HOT 1
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from era3d.