AI paintings of Chinese landscapes pass as human-made 55% of the time, research by Princeton student shows


An AI algorithm that creates Chinese landscape paintings was shown to fool human evaluators 55% of the time, beating other AI art. The algorithm was trained on nearly 2,200 high-quality Chinese paintings, but its own art was made from original sketches. — SCMP

Artificial intelligence has already proven capable of producing music and novels, but now a student at Princeton University has shown that it can also create Chinese landscape paintings that are good enough to fool human evaluators – sometimes.

As part of her undergraduate research, Alice Xue studied whether a machine could pass a Visual Turing Test by producing images that people cannot tell were made by a machine. Xue trained an algorithm using 2,192 traditional Chinese landscape paintings collected from art museums. The resulting AI-generated paintings were mistaken for being made by humans 55% of the time.

Xue graduated earlier this year, but the research was just made public in November when published on the online scientific paper repository arXiv. It was first reported in the AI industry publication Synced this week.

The AI algorithm produced the artwork using a machine learning framework model called Generative Adversarial Networks (GANs). GANs have previously been used for turning photographs into paintings and creating other types of art, some of which have even been sold at auction.

Xue’s algorithms used 2,192 traditional Chinese landscape paintings collected from art museums to generate its own paintings. Picture: ArXiv/Alice Xue

One AI-generated painting from the French art collective Obvious sold for US$432,500 (RM1.75mil) at a Christie’s auction in 2018. Like Xue’s AI-generated art, that painting of a man with blurred, indistinguishable facial features was generated by an algorithm trained on thousands of other paintings. In this case, it was fed 15,000 portraits from between the 14th and 20th centuries.

Xue’s research was also original in its choice of artwork. “Most GAN research focuses on Western art,” Xue wrote in the paper. She said this is why she chose to focus on Chinese landscape paintings, which are “just as aesthetically meaningful as Western art”.

For the Visual Turing Test, Xue surveyed 242 participants. The selection process and the participants’ knowledge of art was not specified in the study. They were each asked whether they thought a painting was made by a person or computer and how certain they were on a scale of one to 10.

Xue said her algorithm is also unique in its ability to create artwork from beginning to end. The traditional GAN-based method of creating artwork tends to produce unoriginal artwork, according to Xue, because it relies on inputs like pre-prepared sketches.

For this research, Xue used a version of the framework called Sketch-and-Paint GAN (SAPGAN). This trained the algorithm to make its own sketches of the landscape’s mountains and rivers before painting in the details.

The results, according to Xue, help lay the groundwork for future machine-made artwork that is truly original. – South China Morning Post

Article type: metered
User Type: anonymous web
User Status:
Campaign ID: 18
Cxense type: free
User access status: 3

SCMP , algorithm

   

Did you find this article insightful?

Yes
No

Next In Tech News

ToonMe is making waves on Twitter – how wary should we be?
EU fines video game firms for blocking cross-border sales
WiFi 6E: what difference will it make for me?
IT loophole lets UK public book vaccine appointments - Evening Standard
Google asks U.S. judge to move states' antitrust lawsuit to California
Factbox: How Facebook, Twitter and others are girding for inauguration threats
Pinduoduo worker deaths reignite debate on 996 and the dark side of China tech’s overwork culture
Tencent seeks video breakthrough after US$240bil lite app hit
Marketing, sports data apps join fight against pandemic
World Wide Web inventor opposes Australia’s news payment plan

Stories You'll Enjoy