Innovative joint experiences between art and technology allow a whole new range of works and novel relationships between viewer, art and artist.
DAMO Academythe research arm of Chinese conglomerate Alibaba, has used artificial intelligence to make ancient Chinese portraits blink, smile and talk, as part of an initiative to bring cultural classics to life with digital technology.
DAMO collaborated with a public library in Hangzhou, the company’s hometown, to animate a dozen paintings, including “Han Xizai’s Nighttime Revelry” by artist Gu Hongzhong and illustrations for “Dream of the red camera”, by Cao Xueqin.
“With AI technology, we can see how historical figures’ facial expressions change as they recite poetry,” said Liang Liang, deputy director of the Hangzhou Public Library, commenting on the collaboration, which the library co-initiated to celebrate the This year’s Mid-Autumn Festival in China.
“We also invite Hangzhou residents to participate and record voiceovers, making this a collaborative cultural event that brings ancient texts and paintings to life,” Liang noted.
The company plans to continue leveraging AI to preserve cultural heritage and classical texts.
This project explores the possibilities of connecting technology with the cultural heritage of a country, bringing old books and paintings to life. It is a new way to present valuable historical data.
Technology as a growth channel for independent art
The process
The DAMO Academy Vision Lab, the same team behind Alibaba’s virtual avatar technology and Pailitao image search tool, spearheaded the project. Lab engineers developed an image analysis algorithm that accurately detects facial features and key points of historical figures in paintings.
They also used videos of Hangzhou residents they recorded for the project, modeling the changes in their movements. DAMO used its AI models to process this mass of information and transform the figures in the still paintings into realistic, moving characters.
Tang Mingqian, an AI engineer at DAMO Academy, said the biggest challenge was making sure his characters really captured the spirit of the original artwork.
To do this, the team reviewed hundreds of thousands of traditional portraits to examine the faces from different angles. This helped make the head movements in the portraits more multidimensional and lifelike.
In addition, they refined image quality and developed an edge-aware attention network module that smoothed out motion and helped prevent shake.
Immersive works, the intersection between technology and art
Other uses of AI in art
Bringing old portraits to life is not the only use of Artificial Intelligence. But it can also be the answer to achieve the conservation of hundreds of thousands of years of history and culture.
Thus, the DAMO academy also launched a project to digitize Chinese classics with the UC Berkeley Library, Sichuan University, the National Library of China and the Zhejiang Library.
The project aims to convert scanned images of ancient books into text for online readers, including using AI and optical character recognition technology to help decipher ancient Chinese characters, which can be complex, with different variants. and written forms.
AI also helps reconstruct works of art that have deteriorated over the years. Thus, the work can be assembled using the same color palettes, following the same patterns and helping to ensure that all the details of it are as exact as they were. AI avoids human error, which has so often made a work go viral for the wrong reasons.
As you can see, technology is not only used to push the boundaries of what is artistically possible, but also helps us look into the past, and make old pieces have a new life in the digital world.
Add Comment