Comments (1)
voeykovroman
commented on May 25, 2024
We use sigmoid to squeeze output values into range between 0 and 1. These values are relative disparities between left and right images and need to be multiplied by image size to calculate, for example, depth of some point on the image.
Moreover we use 0.3 as a constraint to get some reasonable values as the original implementation suggested.
from monodepth-pytorch.
Related Issues (20)
- This code is able to reproduce similar results to those in the original paper HOT 16
- RMSE keeps increasing after 25 epochs training while disparity prediction looks fine
- Question about how to get the original image HOT 1
- How do you compute L-R Consistency HOT 1
- Difference between ResNet50_md and ResNet model
- why the disp_gradient_loss, lr_loss is zero and the total loss is not getting converging?
- test my data HOT 2
- How can I get deep information? HOT 2
- disparity map error HOT 1
- [Question] How to get the metric disparity value? HOT 3
- [Feature requested] Support for depth-separable(depth-wise) backbones such as MobileNetV2, EfficientNet... HOT 1
- Can anyone provide the download URL of pretrained model 'monodepth_resnet18_001_cpt.pth' in this repository? HOT 6
- Why horizontal flip HOT 1
- Key already registered with the same priority: GroupSpatialSoftmax HOT 1
- Does this work on stereo image pair (not video) HOT 3
- How to interpolate the sparse gt depth map？ HOT 1
- What is the loss value of the trained model？ HOT 1
- Training data with monocular HOT 1
- datasets HOT 1
- How can I generate depth maps for other image datasets using the code here？
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 monodepth-pytorch.