Comments (2)
Hi, I can think of two solutions for your purpose:
-
Feed the original two images into the network and you'll get the disparity map w.r.t the left image. Then you can unproject the image coordinates of left image to 3D space, transform them according to the relative camera motion, and project them into the right image. You'll get the disparity/depth map of the right frame, but the density might not be 100%;
-
Swap the left and right images, and flip them horizontally at the same time. Feed the processed image pair to the network, and flip the prediction at last. In that case, you'll get a dense disparity map of the right image.
from hd3.
will try that. thanks for the quick reply!
from hd3.
Related Issues (20)
- How to genereate the confidence map? HOT 1
- Regarding feature map resolution from DLA encoder HOT 3
- Discrepancies model_zoo models vs. paper HOT 1
- output tensors differ when parsing exactly same input HOT 1
- error when run train.sh HOT 8
- Number of epochs and training time HOT 5
- Some Results and Questions HOT 1
- Extra minus HOT 7
- Cityscapes pre-trained model? HOT 1
- Some layers have no gradient when backward.
- Error running pre trained model HOT 1
- Question about how occlusion is handled HOT 2
- Large EPE error for stereo task HOT 3
- replace FunctionCorrelation cupy GPU HOT 4
- Inference error HOT 4
- Convert output disparity maps to depth HOT 6
- Can you give me a link to the web disk of the pre training model?
- Question about pretraining in stereo task HOT 5
- Question about cost volume
- Error in _prob2cornerflow() function HOT 3
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 hd3.