vijishmadhavan / skindeep Goto Github PK
View Code? Open in Web Editor NEWGet Deinked!!
License: Apache License 2.0
Get Deinked!!
License: Apache License 2.0
fastai==1.0.61
fastcore==1.3.27
numpy==1.20.3
torch==1.10.0+cu113
torchvision==0.11.1+cu113
@vijishmadhavan This project is helpful for me, but I don't know how the dataset looks like, does it look like this:
path_hr = Path('image without tattoo')
path_lr = Path('image with tattoo')
Thanks for your kind reply.
This is awesome, thank you for putting this together. If I wanted to extend this basic idea to other use cases (i.e. removing a logo from a car) how would I go about that? I can get both the "before" and "after" pictures of the car for training.
The size of the original image is 9461420, and the predicted image is 400400.
I thnik the size of the two images shold be the same for the best. Could you fix it?
The jpg's website: https://pyxis.nymag.com/v1/imgs/ed6/c4f/367aedb908ac63e80936c174423cc4ac57-26-adam-levine-tattoos-005.2x.h473.w710.jpg
Hi, I've finally gotten this working on my pc, and it's good, not great.....pretty impressive though.
the image sizes are bad.......if you put in a portrait photo, the ideal situation would be not a 400px X 400px square.....we'd want ie the same image size, with the tattoos removed, in portrait / landscape etc.......where abouts in the code do we change this, or is this a function of the actual skindeep2.pkl?
How do we add our own data to the skindeep2.pkl , is this possible or do we have to start off with another one of our own? How do we make a dataset if we already have the images ready to go
it occasionally removes ie peoples shoes, or eyes, or half their face (no tattoos), is this being worked on? It's very cool though, I just thought I'd see if the developer would be available to maybe give us abit of help if we wanted to try to improve this......I work for hours removing tattoo's with photoshop, and even if I got this to ie do a layer to add back into photoshop, to help my work flow.....
Thanks in advance
How many dataset have you used for training the model ?
What is the image size? Do we have to resize image to 64 ? Is it possible for you to share your dataset?
Here is the stacktrace for failing step 3, it can't proceed later
!pip install -r colab_requirements.txt
Collecting fastai==1.0.61 (from -r colab_requirements.txt (line 1))
Using cached fastai-1.0.61-py3-none-any.whl (239 kB)
Collecting numpy==1.17.2 (from -r colab_requirements.txt (line 2))
Using cached numpy-1.17.2.zip (6.5 MB)
Preparing metadata (setup.py) ... done
Collecting pandas==1.1.2 (from -r colab_requirements.txt (line 3))
Using cached pandas-1.1.2.tar.gz (5.2 MB)
error: subprocess-exited-with-error
× pip subprocess to install build dependencies did not run successfully.
│ exit code: 1
╰─> See above for output.
note: This error originates from a subprocess, and is likely not a problem with pip.
Installing build dependencies ... error
error: subprocess-exited-with-error
× pip subprocess to install build dependencies did not run successfully.
│ exit code: 1
╰─> See above for output.
note: This error originates from a subprocess, and is likely not a problem with pip.
would you share your dataset please?
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.