2019 fellow: Justin Kelly

Digital image project: biblio.ai

During his fellowship, Justin will use state-of-the-art computer vision technologies from the likes of Google and Amazon to enrich and bring new depth and possibilities to the Library's digital image collection, in particular, the Image pool digital picture library.

Justin will look at how cutting-edge computer vision technologies can be used to enrich the collection's metadata, including text recognition of all possible text within images, and extensive object recognition such as automatically cataloguing all objects, activities, people, gender etc.

All of these enrichments would be available to be reingested into the Library's various search and discovery systems, as well as the Image pool, via the open source image AI tool biblio.ai.

Justin Kelly is a Data Engineer at Swinburne University in the Business Analytics department, focusing on data warehousing, machine learning, visualisation, computer vision and cloud technologies. These skills combined with his years of experience with library systems provide a unique opportunity to use Justin's understanding of library technologies with cutting-edge cloud technologies that are underused but of extreme importance in the library space.