AI-nimation – Innovations in Computational Creativity.
Created by Joshua Gould, Prepared for the Digital Media Innovation (Summer 2018) course at Rutgers' School of Communication & Information as a part of the Graduate Program in Digital Media & Communication.
Interactive Presentation: https://drive.google.com/file/d/1KBheC_FghDL5lDiJwzf7VsSIC5b8Dz7j/view?usp=sharing
Preface from project proposal:
In the information age, developments in artificial intelligence have unlocked new forms of creativity and content in media. AI, or artificial intelligence, is a truly phenomenal and transformative technology. Cognitive computing has unlocked our ability to extract insights and actions from big data, enables cars to drive themselves and can even create flawless works of art like Rembrandt. We are witnessing a Cambrian explosion of artificial intelligence that ushers advancements in virtual agents’ visual perception and their application towards new art forms and tools. How can we leverage machine learning and artificial intelligence to create in new ways?
This proposal outlines the basic concept, project timetable and viability of a media prototype entitled: AI-nimation. Conceptually, AI-nimation would be a machine-learning tool, leveraging algorithms and neural networks programmed to scan specific visual data, capable of generating animations (or collages) from text-driven descriptions. Similar to functions of computer vision and pattern recognition, AI-nimation would be able to analyze novel text input, or plain-language, and through a database of visual elements create a temporal-layout of mentioned characters and objects. In laymen terms, AI-nimation would be able to intake general scene captions, containing character, prop and location descriptions, and create computer-animated scene videos. In theory, this tool truly enables machine learning and artificial intelligence to design and create.
Created by Joshua Gould, Prepared for the Digital Media Innovation (Summer 2018) course at Rutgers’ School of Communication & Information as a part of the Graduate Program in Digital Media & Communication.
Interactive Presentation: https://drive.google.com/file/d/1KBheC_FghDL5lDiJwzf7VsSIC5b8Dz7j/view?usp=sharing
Preface from project proposal:
In the information age, developments in artificial intelligence have unlocked new forms of creativity and content in media. AI, or artificial intelligence, is a truly phenomenal and transformative technology. Cognitive computing has unlocked our ability to extract insights and actions from big data, enables cars to drive themselves and can even create flawless works of art like Rembrandt. We are witnessing a Cambrian explosion of artificial intelligence that ushers advancements in virtual agents’ visual perception and their application towards new art forms and tools. How can we leverage machine learning and artificial intelligence to create in new ways?
This proposal outlines the basic concept, project timetable and viability of a media prototype entitled: AI-nimation. Conceptually, AI-nimation would be a machine-learning tool, leveraging algorithms and neural networks programmed to scan specific visual data, capable of generating animations (or collages) from text-driven descriptions. Similar to functions of computer vision and pattern recognition, AI-nimation would be able to analyze novel text input, or plain-language, and through a database of visual elements create a temporal-layout of mentioned characters and objects. In laymen terms, AI-nimation would be able to intake general scene captions, containing character, prop and location descriptions, and create computer-animated scene videos. In theory, this tool truly enables machine learning and artificial intelligence to design and create.
