Examining the Affect of Artificial Intelligence in Museums

Brendan Ciecko, Cuseum, U.s.a.

Abstract

Artificial Intelligence. Information technology's a concept that holds lots of hope, generates endless buzz, and is starting to brand its way into everyday life. In 2015, artificial intelligence went mainstream, and undoubtedly, in 2016, we will begin to see an increment in experimentation within the cultural space. In this presentation, we'll explore some of AI's about powerful uses related to machine learning and its bear on on galleries, libraries, archives, and museums in the areas of collections, ticketing, and attendance information. We'll also examine auto vision; a computer's power to empathize what it is seeing. Machine vision can be used to inspect and clarify images. Imagine being able to classify all of your visual objects with the flip of a switch (actually, a few lines of code). We'll explore existent examples of auto learning on the following topics: -Identifying subject matter -Exacting colour composition -Sentiment analysis -Text/character recognition -Recognizing similarity and patterns -Art hallmark Machine learning and vision are very powerful tools and are more accessible than ever before. In the hands of museums, these technologies will inevitably lead to interesting discoveries, rich data, and new paths into your collection.

Keywords: Artificial Intelligence, AI, Car Learning, Car Vision

Artificial Intelligence. It's a concept that holds much promise , generates countless buzz, and is starting to make its mode into everyday life. "In 2015, bogus intelligence went mainstream," (Time, 2015) and undoubtedly, in 2017, we volition brainstorm to meet an increase in experimentation within the cultural space.

We volition explore several of AI's most powerful uses related to machine learning and machine vision focusing on its impact on galleries, libraries, archives, and museums.

What is machine learning?
"Car learning is a method of data assay that automates analytical model edifice. Using algorithms that iteratively acquire from data, machine learning allows computers to find subconscious insights without being explicitly programmed where to look." (SAS, 2016)

Motorcar learning'southward affect on collections
It comes as no surprise that museums have tremendous amounts of information. Strides accept been made over the past decade towards structuring collections' data and making it available for the public to access and experiment with. While still highly untapped, this valuable metadata holds power and yields interesting ways to clarify collections, objects, and creators in new ways. Simply, it likewise requires significant resources, tools, time, and expertise.

In an ideal world, GLAM (galleries, libraries, archives, and museums) collection data would exist structured and well classified, merely given that "more than 90 percentage of (enterprise) data is unstructured, human-generated and sourced from various disparate entities" (IDC, 2015) nosotros tin assume that museum collection data would benefit from some clean-up, maybe fifty-fifty an overhaul.

Could AI come to the rescue, even helping museums make new discoveries virtually their collections? Those working with the museum'due south collections' management system could "train" a system to finer clean-upward, classify, and further understand their data.

Pointing to one big calibration initiative,
machine learning has become a recurring theme amid the Eu's digital platform for cultural heritage, Europeana 's Search Strategy , published in 2016.

Practice you want to quickly run a
sentiment analysis  beyond the title and didactic text of every object in the collection? Yous can–and it'due south becoming exceedingly easy to do with the tools that are currently bachelor.

Run into how three museums take used sentiment analysis:

  • SFMOMA: Sentiment Analysis ( John Higgins, 2015 )
  • Carnegie Museum of Fine art: Gulf Belfry Project
  • Tate: Diving into the Museum's Social Media Stream ( Elena Villaespesa, 2013 )

Car learning's impact on ticketing and omnipresence
Imagine taking those massive sets of ticket and visitor traffic data and using AI to look for clear correlations betwixt them and social media activeness, weather, advertising spending, and other variables.

Research at Pennsylvania State University has investigated methods of predicting omnipresence every bit outlined in the report,
"Who Volition Nourish? – Predicting Event Attendance in Effect-Based Social Network."

It's feasible to say that museum departments could find new and insightful information that could be used to make predicting crowd menstruation, allocating staffing resources, and overall planning more efficient.

Auto learning'due south impact on membership and fundraising
Pattern recognition could easily assistance museums identify members who are most likely to renew, upgrade, or lapse. New tools can assist evolution teams on their fundraising campaigns by deciphering trends, navigating through the social graph , and automating aspects of the donor outreach.

Although relatively new to the market, software companies such as Gravyty and Affectly accept used some of the aforementioned techniques to assistance nonprofits fundraise more finer.

Machine learning's impact on due east-commerce
Major e-commerce sites like Amazon, eBay, and Zappos have been using recommendation and personalization engines for as long as anyone can remember. By analyzing your beliefs, i.e. pages you lot visit, products you look at, and categories you explore, online retailers make recommendations to provide a more personalized feel for each visitor.

Major museum online stores such as those of
The Met , MoMA , and dozens of others already utilise recommendation engines. On the horizon is the mass concept of conversational commerce. Chris Messina of Uber said "2016 will be the year of conversational commerce." (Messina, 2016)

What is motorcar vision?
Machine vision is the power for a computer to understand what it is seeing.

"We're going from computers with cameras, that take photos, to computers with eyes, that can encounter"
– Benedict Evans , Andreessen Horowitz

Back in 2014, the Museum of Arts and Design in New York hosted a panel examining the "Cultural Impact of Computer Vision" from the eyes of artists. Flash forward to the nowadays, and we volition take a look from the perspective of museums.

Touch on of machine vision on identifying subject matter
Auto vision has become advanced enough to find the subject thing and objects depicted in an paradigm. What is depicted in this painting, photo, video, or sculpture?

Figure 1: image of "The K Canal in Venice from Palazzo Flangini to Campo San Marcuola" by Canaletto, J. Paul Getty Museum

Using Google Vision API we tested Canaletto 'southward The Grand Canal in Venice from Palazzo Flangini to Campo San Marcuola located at the J. Paul Getty Museum in Los Angeles. (meet figure 2)

The results were acceptable and positive. The four terms returned ( watercraft rowing , rowing , gondola , and painting ) were all accurate descriptions of the subject affair and objects.

Figure ii: image of Terminal running script to analyze the aforementioned Canaletto painting.

There is however a ways to become with object nomenclature but information technology'due south worth noting that the more you "train" a car vision engine, the more accurate it becomes.

Museums such as the Harvard Art Museums , Minneapolis Arts Museums , Norwegian National Museum are amongst the first to experiment with this approach and share their findings publicly.

Machine vision's touch on sentiment assay
If there are unobstructed human faces in an image, auto vision can exist used to make up one's mind the emotional state of those portrayed past analyzing the facial characteristics.

To put this process to the test, we ran a few portraits through the Emotion API of Microsoft Cognitive Services . (run into figures three-5)

Effigy 3: Image of "Bosom of a Laughing Young Man" (1629) by Rembrandt (circle of), Rijksmuseum.
Figure 4: Image of "Femme aux Bras Croisés" (1901) past Pablo Picasso, Individual drove.
Figure five: Image of "Self-Portrait" (1912) by Otto Dix, Detroit Establish of the Arts.

Machine vision's impact on text/grapheme recognition
The ability to easily extract text from every object in your collection has been possible for many years. The tool ordinarily known as " optical character recognition " has recently become more accessible and faster to use via deject APIs.

Figure 6: Image of "California Grapeskins" (2009) past Ed Ruscha.

While this might non be admittedly necessary for pieces past Lawrence Weiner (as the title and text displayed in his works are usually the same), this role's greatest value could come from extracting text from written documents (historical letters, etc.) then that it's searchable and easy to classify.

In this Ed Ruscha piece titled California Grapeskins, the full text can be successfully extracted, providing additional information that may not exist available in its drove information record. (see figure half dozen)

Machine vision'south touch on on exacting color composition
Color limerick is one meta-tag that you lot are unlikely to find in most museum collections' databases. Running an object'southward image through a figurer vision tool can extract and output data related to its colour clusters, partitions, and histogram data.

Cooper Hewitt, Smithsonian Design Museum and Google Arts & Culture  have implemented this process to extend a new approach to discovery. (see figure seven-8)

Effigy seven: Screenshot of Cooper Hewitt's collections website where visitors can scan objects by color.
Effigy 8: Screenshot of Google Art & Culture'due south app where visitors can browse objects by color.

Car vision's impact on recognizing similarity and patterns
Are there other works in your collections that are very like, not just on subject matter, only visual limerick? A computer can see these relationships and quantify the differences and similarities.

For example, these ii Clyfford Nonetheless "replica" paintings are slightly different, 5.58% to exist exact. (run across figure nine)

Figure 9: Epitome of Clyfford Nonetheless paintings (L to R) PH-225, (1956). Oil on canvas. Collection of the Modernistic Art Museum of Fort Worth; PH-1074, (1956–9). Oil on sheet. Clyfford Still Museum © City and County of Denver

I was personally inspired to uncover this later on visiting the Clyfford Still Museum in Oct of 2015 for Echo/Recreate,  a fascinating exhibition in its own correct. The museum's director of digital media, Sarah Wambold, has also written about this concept in an article titled "Twinsies! " (Wambold, 2016)

Walking into the Impressionism Gallery at the Museum of Fine Arts, Boston , you'll detect these two paintings by Claude Monet, side-by-side. According to estimator analysis the 2 works are 96.81% similar. (run into figure ten)

Effigy 10: Image of Claude Monet paintings (L to R) Water Lilies, 1905. Oil on sail; H2o Lilies. 1907. Oil on sheet. Collection of Museum of Fine Arts, Boston.

Auto vision's touch on art authentication
Dorsum in 2008, PBS NOVA covered the example of computers helping distinguish forged art from original masterpieces . This project was in cooperation with the Van Gogh Museum and challenged computer scientists to build tools to analyze brush strokes and identify forgeries. (come across figure 11)

Figure 11: Image of brush stroke analysis by calculator program.

Contempo revelations: TATE Britain IK Prize
In September 2016, bogus intelligence was the cadre topic of a museum exhibition and project at the TATE United kingdom of great britain and northern ireland . The winner of the 2016 IK Prize utilized various aspects of machine vision, such as subject affair identification, composition, and facial recognition. (see figure 12)

Figure 12: Images of photos and painting from Tate IK Prize. From left: Eduardo Munoz/Reuters. Stephen McKenna, via Tate.

In response to the project and exhibition, The New York Times published the story "Artificial Intelligence as a Bridge for Art and Reality" with voices from the museum customs: "James Cuno, president of the J. Paul Getty Trust and an evangelist for the employ of technology by art historians, assessed 'Recognition' as a 'well-meaning and an interesting experiment.' Then he added, 'It shows that we are in the early on stages of the evolution of this technology and that in that location's nevertheless a long manner to go.'"

Another notable position found in artnet News ' " Art World Predictions for 2017 " stated  that "the burgeoning field of Artificial Intelligence will finally figure out curating" and pointed to a project called HUO 9000. Without question, we tin expect more than projects similar this to emerge that farther challenge the status quo of fine art curation.

Determination
Artificial intelligence is existence lauded as "the future." There is untapped value to be unleashed across sectors seeking its commercial, scientific, and educational potential. With machine learning and vision tools more attainable than ever earlier, museums have the opportunity to innovate and optimize in areas that were previously too costly or resources prohibitive to pursue.

Regarding broader applications of AI, we must acknowledge that creative bots are already creating paintings, writing screenplays, and composing music. In the futurity, will AI write object labels, script audio guides, and assist with interpretation? Should we let machines to do this?

Stephen Hawking predicts "computers will overtake humans with AI inside the next 100 years. When that happens, we need to make certain the computers have goals aligned with ours." This may sound ominous, but we can be (almost) sure that museums and cultural institutions will accept mankind's best interests in mind.

This is just the get-go.

References

Ciecko, Brendan. (2016) Exploring Artificial Intelligence in Museums . last updated February 25, 2016. Consulted February 2017. http://web log.cuseum.com/post/139971318568/exploring-bogus-intelligence-in-museums

Ciecko, Brendan. (2016) 6 Ways that Machine Vision can Help Museums . last updated March ten, 2016. Consulted February 2017. http://blog.cuseum.com/post/140786158798/half-dozen-ways-that-machine-vision-can-help-museums

Davis, Ben, Artnews (2017) Google Sets Out to Disrupt Curating With "Machine Learning" . last updated January 14, 2017. Consulted Feb 2017. https://news.artnet.com/art-world/google-artificial-intelligence-812147

Dobrzynski, Judith.
(2016). Artificial Intelligence as a Bridge for Fine art and Reality. New York Times. https://world wide web.nytimes.com/2016/10/thirty/arts/pattern/artificial-intelligence-every bit-a-bridge-for-art-and-reality.html

Frankel, S., & Hammond, K. (2016). five Predictions for Artificial Intelligence in 2016. Time Mag. http://time.com/4175663/5-predictions-for-bogus-intelligence-in-2016/

Hill, T., Haskiya, D., Isaac, A., Manguinhas, H., & Charles, V. (2016) Europeana Search Strategy. last updated April 23, 2016. Consulted February 2017. http://pro.europeana.eu/files/Europeana_Professional/Publications/EuropeanaSearchStrategy_whitepaper.pdf

Higgins, John. (2016) Sentiment Analysis . last updated February 2015. Consulted February 2017. https://www.sfmoma.org/read/sentiment-analysis/

Markoff, John. (2015). A Learning Accelerate in Bogus Intelligence Rivals Human Abilities. New York Times. https://www.nytimes.com/2015/12/xi/science/an-advance-in-bogus-intelligence-rivals-human-vision-abilities.html

Messina, Chris. (2016) 2016 will be the twelvemonth of conversational commerce . last updated Jan xix, 2016. Consulted February 2016. https://medium.com/chris-messina/2016-will-be-the-year-of-conversational-commerce-1586e85e3991

Museum of Arts & Design . (2014) Cultural Impact of Computer Vision . terminal updated Nov 2014. Consulted February 2016. http://madmuseum.org/events/cultural-affect-computer-vision

SAS . (2016) Car Learning: What information technology is and why it matters . last updated March 2017. Consulted Feb 2017. http://madmuseum.org/events/cultural-affect-computer-vision

Villaespesa, Elena. (2013). "Diving into the Museum's Social Media Stream. Analysis of the Visitor Experience in 140 Characters." In N. Proctor & R. Cherry (eds). Museums and the Spider web 2013. Silver Spring, Doctor. Consulted Feb, 2016. http://mw2013.museumsandtheweb.com/paper/diving-into-the-museums-social-media-stream/

Wambold, Sarah. (2016) Twinsies! . last updated January, 2016. Consulted Jan 2016. https://clyffordstillmuseum.org/twinsies/

Westvang, Evan. (2016) Deep Learning at the Museum . last updated January 2016. Consulted January 2016. http://bengler.no/web log/deep-learning-at-the-museum

Zhang, Ten., J. Zhao, & G. Cao . (2015) Who Will Attend? – Predicting Consequence Omnipresence in Event-Based Social Network. Consulted February 2017. http:// ieeexplore.ieee.org/iel7/7263115/7264280/07264306.pdf


Cite as:
Ciecko, Brendan. "Examining the Bear on of Artificial Intelligence in Museums." MW17: MW 2017. Published February 1, 2017. Consulted .
https://mw17.mwconf.org/paper/exploring-artificial-intelligence-in-museums/