# Wolfram Summer SchoolFounded in 2003

17th Annual Wolfram Summer School, held at Bentley University June 23–July 12, 2019

Alumni

# Sabrina Giollo

Science

Class of 2016

## Bio

I graduated in mathematics from the University of Padua. I used Mathematica for some projects, labs and lectures about prime numbers, fractals, coverings, cryptography and dynamical systems. I've always been impressed by the power and usefulness of Mathematica and Wolfram|Alpha. My bachelor's degree thesis was about iterated function systems and fractals, which I generated with Mathematica. My master's degree thesis was about optimization, operative research and variational inequalities. In particular, I tried to modify and improve an existing algorithm called DIRECT for finding more accurate solutions of variational inequalities. I've just finished an internship for an IT company, during which I developed software for project management and workflow.

I like creating handmade jewelry, and I'm a church volunteer educator.

## Project: Image Colorization

The aim of my project is to build a neural network that could be able to colorize grayscale images in a realistic way. The network is built following the article [1]. In this paper, the authors propose a fully automated approach for the colorization of grayscale images, which uses a combination of global image features extracted from the entire image and local image features computed from small image patches, in order to colorize an image automatically. Global priors provide information at an image level, such as whether the image was taken indoors or outdoors, whether it is day or night, etc., while local features represent the local texture or object at a given location. By combining both features, it is possible to leverage the semantic information to color the images without requiring human interaction. The approach is based on convolutional neural networks, which have a strong capacity for learning. The model is trained to predict the chrominance of a grayscale image using the CIE L*a*b* colorspace. A nice property of predicting colors is that training data is practically free: any color photo can be used as a training example.

## Reference

[1] S. Iizuka, E. Simo-Serra and H. Ishikawa, "Let There Be Color!: Joint End-to-End Learning of Global and Local Image Priors for Automatic Image Colorization with Simultaneous Classification," in Proceedings of ACM SIGGRAPH 2016, New York: Association for Computing Machinery, 2016.

## Favorite 3-Color 2D Totalistic Cellular Automaton

Rule 552697707

I chose rule number 552697707 starting from a single black (blue) cell in the center.