Comments (2)
Hi Menpinland,
- Good question! Thanks for pointing this out. The main reason why our model does not produce quite good relighting in this case is that our model are based on low level spherical-harmonics, which is designed to render images with very ambient lightings. Since you are using a lighting map with a very strong source (the sun), it will drasically affect the approximation from the SHs, and thus causing the artifact. An easy solution to that is to replace the SH-based rendering to other methods, such as SG-based rendering or just brute-force, and there're many resources you can look into. In the meantime, you can also try with some more ambient lighting, such as a cloudy outdoor scene, or an indoor scene with ambient lights.
- I'm not quite sure what's the question here. The self.env_lights is used during the rendering at here and thus trained with the back-propagated gradients. Let me know if this answers your question.
from neroic.
Hi Menpinland,
- Good question! Thanks for pointing this out. The main reason why our model does not produce quite good relighting in this case is that our model are based on low level spherical-harmonics, which is designed to render images with very ambient lightings. Since you are using a lighting map with a very strong source (the sun), it will drasically affect the approximation from the SHs, and thus causing the artifact. An easy solution to that is to replace the SH-based rendering to other methods, such as SG-based rendering or just brute-force, and there're many resources you can look into. In the meantime, you can also try with some more ambient lighting, such as a cloudy outdoor scene, or an indoor scene with ambient lights.
- I'm not quite sure what's the question here. The self.env_lights is used during the rendering at here and thus trained with the back-propagated gradients. Let me know if this answers your question.
Thank you for your answer. I will try to get familiar with more lighting models and make more attempts. For the 2rd problem, I have misunderstood some details in the paper and caused confusion.
from neroic.
Related Issues (20)
- Pre-trained Model HOT 4
- Question on Foreground Masks HOT 1
- TypeError: __init__() got an unexpected keyword argument 'every_n_val_epochs'
- Add mesh preview every XX epoch? HOT 1
- Segmentation masks HOT 1
- how to run it with a custom single image HOT 2
- Inconsistent hyperparameters reported in the paper and configuration files HOT 1
- Can NeROIC run on CPU? HOT 2
- Results not as good on custom dataset HOT 2
- Is there any way to export 3D models to run on general 3D software HOT 3
- Help! While running the code, an error occurred. HOT 1
- Is there any model files like obj files generated during training procedure or test procedure? HOT 3
- Training stuck
- How to obtain / export the model generated? HOT 1
- can not train HOT 3
- Error in dataset.py , to fix add additional check in padding function in ./dataset/dataset.py
- how long does this take to run? HOT 1
- geometry.yaml and geometry.yaml in NeROIC HOT 1
- The milkbox in the video I generated rotates very quickly. How can I reduce the rotation speed?
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 neroic.