Historical costume descriptors bridge gap between past and present

The boldest decade in fashion history was arguably the 1980s. Iconic parachute pants, shoulder pads, windbreakers, acid wash jeans, dramatic earrings, spandex leggings — all full of bright, eye-straining colors and patterns. As times change, so do clothes and the words we use to describe them. 

Virginia Tech fashion experts and computer scientists have teamed up to create a common language to describe clothing items in the university's Oris Glisson Historic Costume and Textile Collection that range from centuries old to present day thanks to the University Libraries Collaborative Research Grant for Humanities and Social Sciences. Two Virginia Tech student positions were funded through this grant, allowing them to redescribe the garments, identify new terms, and perform quality control of the descriptions. 

The objective is to compare language from the past with new language that we use today to describe costume artifacts across different time periods. The Oris Glisson Historic Costume and Textile Collection will be digitized and made available to the public, increasing accessibility and access. There are over 5,000 artifacts in the collection, making this a long-term project. The team has described and digitized approximately 200 items so far. 

The collection consists of men's, women's, and children’s garments of all types, including evening wear, nightgowns, undergarments, and a rare 1840s wedding dress. There are accessories such as jewelry, glasses, scarves, and unique hats. Additionally, the team has created a "world closet" that contains items such as kimonos, Chinese skirts, and Korean handbags. The collection’s oldest artifact is a Peruvian burial garment fragment from 1180, although most of the garments date between 1840 and 2010. 

So just how does the team go about creating the common language? 

Dina Smith-Glaviana, director of the Oris Glisson Historic Costume and Textile Collection, oversees the artifacts and the students who redescribes them. They examine them, identify other terms that might be more general, and pull language from the International Council of Museums terminology for costumes. Next, they create the new descriptions to put into their new metadata schema, Costume Core, which means building on existing standards to create a consistent way to catalog and code historic clothing. 

During a typical shift, the students find the list of artifacts to digitize, pull one of those garments from the collection, lay it down carefully on tissue paper, and examine it from top to bottom. Looking at the garment, they research and choose the most relevant terms to describe it. Students also input missing information such as donor name and write a very thick description of the garment. 

“I’ve trained my students to be meticulous and look through an encyclopedia of textiles to make sure they’re identifying the right type of fabric or fabric structure,” said Smith-Glaviana. “They also use the Fairchild's 'Dictionary of Fashion,' which has illustrations of things such as collars, sleeves, skirt types, and much more. The pictures help them find the most relevant term, and it’s a good system to use for people who are not familiar with all of the costume terminology.” 

“I’m a perfectionist, and it’s crucial to the project that all of the details are there,” said Smith-Glaviana. 

The next step is to send the detailed descriptions on notecards to Chreston Miller, data and informatics consultant at University Libraries, where he and his student team oversee the natural language processing, also known as machine learning. There, they compare the cards and predict what Costume Core terms should be used to describe the item.  

“This project is pretty interesting,” said Miller. “I’m a computer scientist, so when I tell people I am working with fashion, they are like, what connection is that? The people in Fashion Merchandising and Design have data they want to work with.” 

Given the challenges of describing artifacts, especially ones of historic nature, Miller said one reason to have a controlled vocabulary is the descriptions that come with an artifact were written in a certain time period, so they have certain phrases for different aspects of the artifacts. “We read it today and we say, 'What does this mean?'” said Miller. 

Graduate student and University Libraries student employee, Madhuvanti Muralikrishnan '23, is assisting Miller on the project while working on her master's degree in computer science and applications. Her role is implementing the algorithms that map a given description to the Costume Core vocabulary.  

“In machine learning, we’re essentially trying to teach the computer to do something that humans do, and that’s pretty complex,” said Muralikrishnan. “We know what multicolored is, but how do you teach that to a program and the intricacies involved in that? That was surprising for me. I read about these things in class, and it’s very gratifying being able to apply what I learned and seeing the result of it.” 

Muralikrishnan said after receiving her graduate degree she wants to work in machine learning, and this project is directly related to her career goals. “I believe that computer science is domain agnostic. So today we’re doing this for fashion. Tomorrow, I can take the same algorithm and do it for medicine or something else. Each domain has its own problems and challenges and that’s been fascinating.”

So far, Miller has received more than 5,000 notecards with descriptions that need digitizing and organizing into a database so everything is accessible and searchable. The notecards start off on paper, then Miller performs optical character recognition on the notecards so the team can digitize the descriptions from the notecards, a crucial step for making the natural language processing possible. 

One item or notecard can take up to 45 minutes to process. The team also uses an online database to share information. “You think, 'OK, color doesn’t seem too hard to describe,'” said Miller. “Well, how many different words of color can you use? Take blue for example. There’s navy blue, baby blue, royal blue, and so on. So you have to think about how you have a control vocabulary for this.” 

Terms are inconsistent over time. “I noticed in a lot of our old records that some of the terminology is not quite accurate or is very specific to the time period in which it was recorded,” said Smith-Glavina. “One thing that has stuck out to me in the collection is the term bloomers. Bloomers was actually a term used in the 1850s to describe the first form of female trousers worn by dress reformers and by bicyclists as a sports costume. But over time, it changed to be ruffly, fluffy balloon-like undergarments mainly for babies — the ruffly underwear that they wear over their diapers that people tend to describe as bloomers.” 

Virginia Tech’s rare fashion artifacts are usually donated by people cleaning out the homes of their parents or grandparents, and they find trunks with many generations of old garments in them. 

“Most of the time they won’t know what the items are and it’s up to us, the experts, to identify what the garments are,” said Smith-Glaviana. “We’ll ask the donors questions like, 'Who do you think wore this?' And that will help us figure out if it’s a female or male, older or younger person, or even a child because sometimes it’s hard to tell with historic clothing. We get a sense of the history from the donor but usually their knowledge is quite limited.” 

Their ultimate goal is to digitize the historic costume collection and make it accessible to the public anyone who’s interested in researching historic costumes. Terms that are accurate, understandable, and consistent allows the average website user or noncostume history expert,  to search websites terms that they understand even when the items they are looking for are very specific. 

“For example, most people know corsets as corsets,” said Smith-Glaviana. “But they don’t realize that throughout much of history corsets are actually referred to as ‘stays’ and they’re two different garments. So depending on the words they search, they may never find what they’re looking for.” 

A collection that was once hidden is now beginning to be searchable by anyone online, falling in line with Virginia Tech’s goals of providing more access to information. Ultimately, the team members hope that their descriptor categorization will be widely adopted by other costume collections, creating a consistent terminology that everyone can use. “It’s a huge vision,” said Smith-Glaviana.