Comments (2)
You probably need to extend the C++/Cuda code for that. Shouldn't be too hard, but it will be very verbose.
Essentially, you go from a 4D cost volume (x, y, shift x, and shift y) to 6D cost volume (you add the z and shift z dimensions). loops must be changed accordingly, as long as tensor expected dims, and border checks.
What you can do though is try to decompose the problem in 2D slices and use the 2D correlation here, only to restack the ( z * shift z) 4D slices together at the end. This can be done with only python, but will be much slower than rewriting it in C++.
from pytorch-correlation-extension.
thank you very much for your advice. I'll try it.
from pytorch-correlation-extension.
Related Issues (20)
- ROCm support?
- Issue with Computing Complete Pixel-to-Pixel Correlation HOT 2
- IndexError: list index out of range HOT 5
- python setup.py install failed HOT 4
- Successfully installed, but unable to import in the code. HOT 4
- error: command 'D:\\Tool\\cuda11.3\\bin\\nvcc.exe' failed with exit code 1 HOT 5
- Hello, my error has not been completely resolved. On the latest version of vscode, once I referenced this package, I couldn't use the debug feature and had to run it directly. But sometimes running it directly throws an error. HOT 4
- Three months ago, I can use the pip command to compile this, but now it's not working. HOT 1
- Compile once and run on other GPU(different computational capacities) HOT 1
- failed to pip install spatial-correlation-sampler
- failed to python setup.py install HOT 1
- pip install fails with "No CUDA runtime is found" despite existing CUDA toolkit installation HOT 1
- some errors with python setup.py install HOT 1
- python setup.py install failed HOT 1
- Hello,I want to ask python setup.py install failed HOT 3
- OSError: CUDA_HOME environment variable not set when python setup.py in Dockerfile HOT 5
- failed to install spatial-correlation-sampler HOT 1
- torch.compile fullgraph=True HOT 6
- Fail to use "python setup.py install" to install spatial-correlation-sampler HOT 4
- Issues with CUDA 11.7 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 pytorch-correlation-extension.