Titulo Estágio
A Modular Framework for Parametric Evolution
Áreas de especialidade
Engenharia de Software
Sistemas Inteligentes
Local do Estágio
CISUC
Enquadramento
Photogrowth [1,2] is an evolutionary system that generates non-photorealistic renderings of images. A painting algorithm, inspired on ant colony approaches, produces emerging renders by simulating artificial ants that travel on a canvas. The trails of the ants are represented by continuous lines, working like paintbrushes. The superimposition and intertwinement of all trails, together with their variation in direction, width, and colour, produce expressive renderings from a given image that would be hard to execute or even imagine.
Each ant has a set of sensory vectors that return the energy value of the surrounding areas. Each sensory vector has a specific direction, length, and weight. The different settings related to the sensory capabilities, life span, reproduction rate, free will, etc. are defined by a wide range of parameters that, when changed, allow the creation of different types of imagery.
Since the number of parameters for controlling the behaviour of the ants and their interdependencies is too large to be manageable by hand, an Evolutionary Algorithm (EA) was adopted to search the space of all possible parameters that control the behaviour of the ants.
Later, to overcome some of the limitations of user-guided evolution, we explored a meta-level interactive art approach where the users could express their intentions and goals through the design of fitness functions. The system also includes a high resolution rendering mode that converts the trails of the ants in vector graphics.
[1] - Machado, P., Martins, T., Amaro, H., & Abreu, P. H. (2014, April). An interface for fitness function design. In International Conference on Evolutionary and Biologically Inspired Music and Art (pp. 13-25). Springer Berlin Heidelberg.
[2] - https://cdv.dei.uc.pt/photogrowth-ant-painting/
Objetivo
The current system is the outcome of several contributions of different programmers with different backgrounds, which resulted in a monolithic piece of software that is becoming increasingly difficult to extend and manage.
The goal of this work is to build a modular Evolutionary Computation framework for parametric evolution and the instantiation of this framework with Photogrowth. In a second stage the framework will be applied to other visual domains and tasks, testing its generality.
The design and development of the framework should follow good Software Engineering practices, ensuring its adequacy, modularity, and extensibility.
Plano de Trabalhos - Semestre 1
- Analysis of the current state of the Photogrowth Application
- Analysis of the Design Patterns that should be used to implement a modular system;
- System Architecture Design
- Implementation of the designed system
- Writing of the intermediate report
Plano de Trabalhos - Semestre 2
- Validation of the implemented system using the Photogrowth domain
- Expansion of the system to other domains
- Experimentation and validation of the system in the selected domains
- Code documentation
- Writing of the final report
- Writing and submission of a scientific Paper
Condições
The student will work in the CMS group of CISUC, with access to all its facilities and resources. A six month scholarship is available for the second semester.
Observações
Strong skills in programming and Software Engineering. Will to communicate in English with other researchers is also important. Other skills of interest include Complex Systems and Artificial Intelligence
Orientador
Penousal Machado & Nuno Lourenço
machado@dei.uc.pt 📩