This I-Corps project utilizes artificial intelligence (AI) principles to detect abnormalities in images - in particular, it focuses on ways to enable visual information extraction. Large collections of art and other images exist in digitized form but there are few automated methods for processing and indexing this data. This project facilitates a wide range of applications from education to online markets because the technology focuses on automated analysis of art and other images based on algorithmic methodologies in artificial intelligence. For example, the project's computational models might facilitate indexing large-scale collections of art and extracting useful information from this massive visual data. In this context, knowledge can be useful to online and traditional galleries, collectors, educators, authenticators and others in making informed decision regarding artwork and artist. While the AI techniques are applicable to a broad range of images, the I-Corps team initially developed prototype computational models for analyzing artworks based on the visual information extracted from paintings. The computational models enable the system to perform perceptual and cognition tasks on digitized art for classification of paintings. AI algorithm can be specialized or tuned and optimized for different domains of fine-art analysis. While initially focusing on customer discovery within artistic contexts, the technology has the potential to be applicable in many different contexts. The I-Corps experience will help the team to better understand the market scope.
|Effective start/end date||9/1/16 → 10/31/16|
- National Science Foundation (NSF)