Nov 29 2018
Cornell scientists and Bloomsbury Publishing are collaborating to assess the advantages of using artificial intelligence (AI) to enhance and optimize the quality and speed of manual indexing for consistent fashion trend forecasting.
Researchers will draw on images and metadata from the Bloomsbury Fashion Photography Archive to explore rapid archiving and image classification through machine learning and AI. (Image credit: Niall McInerney/Bloomsbury Publishing)
The project’s aim is to nurture research connections between the computer vision and fashion world and thereby advance the state-of-the-art in fine-grained visual recognition for fashion and apparel.
Doctoral students Mengyun Shi and Menglin Jia from Human Ecology’s Department of Fiber Science and Apparel Design will examine on images and metadata from the Bloomsbury Fashion Photography Archive to explore fast and smart archiving and classification of fashion images through the application of machine learning and AI.
“Success of this project will open a door to highly reliable trend forecasting, and help the fashion industry respond to changes in consumers’ need and fashion taste quickly,” said Huiju Park, associate professor of fiber science and apparel design, who is co-advising the project with computer science faculty members Serge Belongie and Kavita Bala.
The project is projected to be able to identify particular features of styles and garments and will probably have applications well beyond the Fashion Photography Archive.
“The attribute dataset that this project will create will open new opportunities to recognize fine details in apparel, and will drive new problems and solutions in recognition of fashion and apparel,” Bala said.
Bloomsbury, an award-winning global independent publisher of fiction, nonfiction, children’s, specialist trade, and academic publishing, acquired the archive of over 750,000 images in 2011 and has been manually indexing each image by garment, person, color, and theme for Bloomsbury Fashion Central, a collection of digital resources for the academic educational community.
We are tremendously excited to be in a position to supply a data sample that is sizable enough to assess whether machine learning can improve upon manual indexing.
Kathryn Earle, Managing Director, Bloomsbury Digital Resources.
Subject indexing is the act of classifying or describing a document by index terms or other symbols so as to summarize its content, to specify what the document is about, or to increase its findability. Indexing done by humans is susceptible to error, subjectivity, and knowledge boundaries – something AI could avoid.
We are anxiously awaiting the outcome of Cornell’s research. The Fashion Photography Archive contains priceless images that Bloomsbury has preserved for posterity, and we want to do everything we can to support them online. If AI can improve discovery for researchers and students, we will be thrilled.
Kathryn Earle, Managing Director, Bloomsbury Digital Resources.