Machine Learning applied to Everybody Edits, powered by Keras

This repository contains tasks focusing on prompt engineering for vision models. Each task explores different aspects of image segmentation, object detection, and image generation using advanced machine learning models. Below are detailed descriptions of the tasks and their respective notebooks.

[JOURNAL TIP] 002-IMAGE-ARTIFACT-GENERATION

Explore linear regression models for predicting university admissions and implement a non-linear Radial Basis Function (RBF) regression model for image inpainting.

Deploys LaMa Image Inpainting as a microservice. Part of "WolfPack: Application-Network Co-Design for Edge Resource Provisioning"

A small Neural Network for Image Inpainting, developed for educational purpose in an Introduction to Machine Learning course.

PosePerfect is a Python project that uses generative AI for object pose editing. It detects and segments objects using YOLOv5 and SAM, then modifies their pose with Stable Diffusion inpainting. The project outputs segmented or inpainted images and includes simple setup and usage scripts.

SPL Paper Codes

Completación de Imágenes mediante métodos de EDP numérico.

Code and reference images for the Deep Capsule Prior experiments

StableSAM is an image inpainting tool combining Stable Diffusion and Meta's Segment Anything Model, allowing users to edit images with natural language prompts and precise region selection.

Generate image inpainting with adversarial edge learning

Landscape image inpainting with UNet and Encoder-Decoder

skin retouching and using mouse-handling with OpenCV

Computer Vision

An UWP app with intuitive controls (drag & drop, mouse wheel / click and keyboard usage) to inpaint images.

Image Inpainting using Partial Convolutions