Comments (2)
Hey, good questions.
The only reason for the sigmoid "padding" exists it that helped optimization in some corner cases, like when the learning rate is very large. This is because sigmoids saturate for very large or very small inputs, which causes the gradient to go to zero, which can cause optimization to catastrophically fail. Padding means that the model can emit 0 or 1 without saturating the sigmoid and therefore killing the gradients.
Shrinking the range of the sigmoid is probably going to cause problems in some circumstances, because it means that absolute white and absolute black will both be inexpressible for the model. I'd expect this to drive rgb values to the tails of this sigmoid, which will cause dead gradients.
It makes sense to me that playing with this parameter will affect how "clean" the acc
images look. I'm personally of the opinion that the performance on the background of those Blender objects isn't super meaningful: When the background is a white void, it's equally "correct" to say that distant regions are empty (in which case NeRF's rendering model will fill it in with white if the flags are set accordingly) or to say that distant regions are occupied with dense white stuff. Both answers will produce equivalently good test-set rendering accuracy. If the model can't easily express rgb=1
(which is the cause for a non-padded sigmoid) then it will say that the background is empty, and if it can easily express rgb=1
then it might fill in the background or might not. Whether that is a good or bad thing depends on the application.
from mipnerf.
Thanks for the thorough answer!
It makes sense that shrunken sigmoid saturates and that will cause dead gradients for color, but my main thinking is that having it be padded allows an equivalence between alpha = 0, rgb != 0
, and alpha != 0, rgb = 0
, whereas with normal sigmoid it never reaches 0 except in the limit, thus guiding the model to infer alpha = 0
. I'm experimenting on the D-NeRF dataset with a shrunken sigmoid, and found this to lead to better results than a widened one because it may learn some view dependent components for fully black space in some views. I suspect for just novel views, either one works equally well.
from mipnerf.
Related Issues (20)
- Could you provide the Pre-Trained model?
- Batch_size Can't reduce GPU Memory HOT 5
- Why are we using cylindrical volume representation instead of conical frustum for forward facing scenes? That's because of changing camera's coordinate system to NDC? HOT 1
- Question about mipnerf360 HOT 4
- Question about CPU OOM. HOT 2
- Extracting isosurfaces of Mip-NeRF HOT 1
- Radii computation with and without NDC HOT 3
- Why positional encoding for viewing direction? HOT 3
- a bug about training process
- Why the training time is longer than NeRF HOT 4
- average error matrics in paper HOT 2
- Is it safe to use unnormalized ray directions while sampling points? HOT 3
- Tiny typo in supplemental material HOT 1
- origins and directions issues HOT 2
- scripts/train_blender.sh: line 31: 11426 Segmentation fault (core dumped)
- A confusion about the order of sin and cos in the IPE part HOT 3
- Is dx (or base_radius of cone) all the same for rays of different pixel in a picture? HOT 1
- Use normalized direction vector or not? HOT 1
- How to understand "The variance of the conical frustum with respect to its radius r is equal to the variance of the frustum with respect to x or (by symmetry) y. " HOT 4
- How to get equation (5) 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 mipnerf.