One A-lister is trying to use AI to make art, instead of just inspire it. She is Kristen Stewart, the well-known star for her role in Twilight movies. Who knew her interest was in AI. The actress releases a research paper for her screenwriting debut, Come Swim. The paper was released on ArXiv, an online research repository run by Cornell which publishes papers before they’ve been peer reviewed.
Stewart’s starting inspiration point for Come Swim was one of her own paintings. The paper describes the filmmaker’s experiments with style transfer, a popular use of machine learning that transforms one image into the artistic technique and color profile of another. Stewart and her producers used the technology to turn scenes of Come Swim into the style of Stewart’s own painting. Stewart’s co-authors on the paper are a producer at Starlight Studios and an Adobe employee, whose involvement in the film is unknown.
The paper’s most interesting aspect is its ambition: The team originally tried to tune the algorithm to transfer the sense of emotion in the painting.
“The painting itself evokes the thoughts an individual has in the first moments of waking (fading in-between dreams and reality),” the authors write. “This directly drove the look of the shot, leading us to map the emotions we wanted to evoke to parameters in the algorithm.”
According to the paper, trying to direct the algorithm into producing an artistically satisfying image proved more difficult than expected. The team cropped and added blocks of texture to the input images, so the algorithm would be sure to include those influences more heavily when making the final image.
Come Swim is headed to the 2017 Sundance Film Festival later this month and has been described as “a diptych of one man’s day; half impressionist and half realist portraits.”