paolobezzina / content-blending-and-image-inpainting

This project aims to implement a selection of computer vision techniques that can serve as a foundation for more complex applications and their evaluation.

ARI2129: Principles of Computer Vision for AI - Assignment

The aim of this project is to provide an opportunity to implement a selection of computer vision techniques that can serve as a foundation for more complex applications and their evaluation. Packaged OpenCV functions as well as self implemented implement ones were used.

For this Project, the COTS dataset was used, available for free through the website (www.cotsdataset.info). The COTS dataset presents scenes with an incremental nature.

The first part of this project tackles the problem of content blending across different images.

In the second part of this project, several inpainting techniques were covered together with their evaluation.

Dependencies

• Python 3.8.8
• IPython 7.22.0
• jupyter_client 6.1.12
• jupyter_core 4.7.1
• jupyterlab 3.0.14
• notebook 6.3.0

Installing

To install all above libraries at one go

pip install -r requirements.txt

Executing program

Since both Tasks are in a '.ipynb' format, in order to run they should be opened from any notebook editor however Jupyter is recomended