Comments (5)
Hi!
Thanks for your interest! The training data consists of a few dozens of images with all the stars deleted in PS by hand (there is no other way ensuing high quality of star removal and leaving all other areas of images exactly as they are).
If you want to use it for your particular data, you can clean a few images of your own data and fine-tune existing weights so it works better for your particular star profiles.
Unfortunately, I am not publishing my own training data under any conditions since that's where the huge amount of work went, but I am available to help you as much as I can. Let me know if you are interested and we can discuss things in more details!
from starnet.
Thanks @nekitmm for your help! We discuss it within our team (and probably try to prepare our own images) and I will return with more questions if needed.
from starnet.
Thanks for reaching out, @Amarchuk!
Please do not hesitate to reach out to me with any questions! I am actually very interested in how this will turn out and available for active participation in your project if you want! I am myself in academia and if this will turn into some scientific publication, that will be awesome!
from starnet.
Awesome! What are your research interests?
It seems like for now creation of Photoshop-corrected images is the hardest part of preparation (there can be literally hundreds of stars on the image and I don't know yet if it is possible to remove them). How you manage to do that? Any tricks or advices are highly appreciated.
from starnet.
@Amarchuk, lets take this conversation elsewhere
from starnet.
Related Issues (20)
- Build steps for PI module HOT 1
- Looking to freeze trained data for PI HOT 20
- no support tif files from photoshop? HOT 6
- Trouble processing a mono tif image with Mode='Greyscale' HOT 2
- C++ sources? HOT 4
- Scipy error HOT 1
- Tensorflow issues HOT 4
- Error - Failed to load the native TensorFlow runtime.
- Trying to train and get cannot import name 'toimage', etc.
- no interface in v2 version for MacOS HOT 1
- i want to create with a starmask HOT 1
- Enhancement: I created a Docker image for Starnet HOT 2
- Pixinsight image container and starnet2 plugin issue HOT 9
- GIMP Plugin for StarNet - gimp-plugin-starnet
- Need more detail on training own data
- scipy toimage is deprecated and has been removed HOT 3
- Incompatible with latest version of tensor flow etc HOT 10
- OSX save issue HOT 1
- Crashes after epoch when training (number of columns error) HOT 2
- Training the AI 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 starnet.