akshitagupta15june / akshitagupta15june/Face-X

Cross stitch face pattern finder

Ouverte
#1,569 2 commentaires 0 réactions 0 personnes assignées Voir sur GitHub
enhancement
Langage dominant
Jupyter Notebook
Étoiles
853
Forks
596
Métriques de merge des PR
Aucune PR mergée en 30 j

Description

**Is your feature request related to a problem? Please describe.**
Cross stitching is a form of embroidery that involves creating decorative patterns or designs by sewing X-shaped stitches onto fabric. When you want to create an embroidery of a person you'll have to first create a simplified sketch of it ie...you can't embroide that image with the same exact colors, cause there'll be a lot of them. Take a look at this image, we can see that this image consist of a lot of colors. Its difficult to embroide this person
![obama](https://github.com/akshitagupta15june/Face-X/assets/88505266/46780b2c-7240-4f39-a6fe-2b342b0bae1e)

**Describe the solution you'd like**
The above mentioned problem can be solved by using machine learning ie.. the K means algoritm can be used to reduce the complexity/no0fColors by image quantisation. The number of colors and the desired image size can be given as an input (note that if you have a embroidery frame with a holes then you shall pass in 100 as size, here is an example for size:100 and 50, colors=5)
![100 oba5](https://github.com/akshitagupta15june/Face-X/assets/88505266/20e1a700-4ca8-4795-a64e-18799b05b9b9)
![50 oba5](https://github.com/akshitagupta15june/Face-X/assets/88505266/280b9c82-1a04-4a10-b1be-7735969727b2)

**Describe alternatives you've considered**
I personally think that this is the most effective solution🙂

**Approach to be followed (optional)**
null

**Additional context**
before and after
![obama](https://github.com/akshitagupta15june/Face-X/assets/88505266/f6a89242-64cd-489e-801c-4f432416f57f)
![100 oba5](https://github.com/akshitagupta15june/Face-X/assets/88505266/3b3f9f6b-54bc-4145-a11f-3bd0925bb75a)

Guide de contribution

Ouvrir le guide de contribution

Évaluation

Cette issue n'a pas encore été évaluée.

Recevez les nouvelles issues par e-mail

Un résumé court des issues GitHub adaptées aux débutants.