pixel

Weaving threads of art and humanity

The Museum of Art & Photography in Bengaluru, India is using AI to uncover hidden links between South Asian textiles and art from around the world. 
AI searches the database for artwork that connects to your piece through several streams such as royal & political, similar fabrics, or sacred & ritual.

Textile is more than just fabric. It’s a moment in history that spans cultures and time.

Textiles often capture moments in history and weave works of art that link cultures. Stories can be found in vivid cloth and faded tapestries all over the world. Exquisite fabrics are an integral part of India’s culture, but many global art lovers have yet to discover the bond between these fabrics and other works of art. The Museum of Art & Photography (MAP) in Bengaluru—India’s tech capital—hopes to change that.

MAP launched an educational platform—The MAP Academy—to share the history of South Asian art. Their mission bloomed in 2020 with an online exhibit, making it one of India’s first digital museums. The enthusiastic response made MAP want to find new ways to help art lovers and institutions engage deeper.

We’ve always wanted to be the museum for the future, experimenting and pushing the boundaries of what technology can do in the museum space—not for its own sake, but as a meaningful tool for improving access and enhancing cultural experiences
Abhishek Poddar
Founder & Trustee, MAP

This vision drove them to collaborate with Microsoft and create INTERWOVEN, a new tool that uses AI to connect textiles within MAP’s collection to art around the world. Select a textile within MAP and span over 10,000 years to see a plethora of other pieces that connect to it in surprising ways.

We want to show our audience that art is fun, it is not elitist, and it is for everyone.
Abhishek Poddar
Founder & Trustee, MAP

“While the project began with an idea to explore MAP’s entire collection, one of the artforms that speaks so eloquently to global connections is textiles,” says Kamini Sawhney, MAP’s Director.

At the core of INTERWOVEN is a robust analytical engine powered by Microsoft AI. Computer vision and text recognition software distills images and descriptions of artworks into a massive database. A machine learning algorithm parses digitized information to discern fine patterns that link contrasting or similar works. As new collections are added, the database and connections will grow.

For art historians and curators, INTERWOVEN can help uncover commonalities that inspire new ideas. The collection is uniquely suited to showcase how South Asia’s tradition of textile creation has inspired other cultures. MAP invites everyone to journey through its archives and uncover the unexpected.

“INTERWOVEN reminds us both of the vibrancy of different cultures and how these traditions are shared in a conversation across time and place,” says Brad Smith, President and Vice Chair, Microsoft.

Developed in collaboration with Microsoft’s AI for Cultural Heritage initiative, INTERWOVEN uses machine learning to reveal common themes and shared histories in artistic traditions across cultures.