Comments (2)
What i understand from the code that 'center' and 'scale' are used to make the bounding box so they will predict according with the area of bounding box. If you check again the function of utils.bounding_box, you will see there are initialization of 'center' and 'scale'.
For .t7 i cannot certainly sure about my thought, but this is how i get it. Those .t7 files that existed are used to get information for landmark localization of each images in the dataset that provided (since that dataset is provided from another) and they used 68 points of landmark localization for the face alignment.
I hope my explanation could help you to understand how it works, because i also face the same problem as yours.
The easiest way to run the code with your own image is make new figure to display your own image and use similar gnuplot setup from original dataset. Then, adjust it (scale and center) according with your own images.
Best Regards.
from binary-face-alignment.
@ncuxiaogang The code doesn't directly support this as of this point but it should be straight forward to adapt it. The method expects a bounding box of the face based on which an aproximative scale and center of the face is computed. You can of course also directly try to supply the center and the scale.
I will be updating the code this month at some point to make it easier to run it on your own data.
from binary-face-alignment.
Related Issues (12)
- Visualize the net structure! So complicated! HOT 6
- Training code
- The result of my training is very poor,why?
- conversion from bnn.SpatialConvolution and bnn.binary to nn or cudnn HOT 5
- Very large GPU memory consumption HOT 1
- where is the groundTruth in eval? HOT 1
- AFLWPIFA-val.tar can not download HOT 4
- Error when 'luarocks make' HOT 3
- How to convert to bnn.BinarySpatialConvolution from cudnn.SpatialConvolution.cu after training HOT 1
- Visualize the bounding box HOT 3
- How convert this model to pytorch? HOT 4
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 binary-face-alignment.