Google DeepMind and filmmakers used AI to recreate a couple's unrecorded past in 'Love, Rendered,' a short film exploring memory loss and storytelling. The film follows Burt and Ethelle Shatz, a couple married for over 70 years, as they navigate Burt’s cognitive decline.
Google DeepMind engineer Michael Chang said, "Memory loss is personal to me. My grandfather suffered a stroke and memory loss before he passed." This personal connection drove Chang to explore how AI could help connect with family through storytelling.
The project involved restoring black-and-white photos of Burt and Ethelle from their youth using generative models. These restored images helped ensure the subsequent recreations remained faithful to their subjects.
Working alongside Primordial Soup, the team used performance capture models to map Burt and Ethelle’s present-day micro-mannerisms. Mapping those traits onto their younger likenesses brought the recreation to life.
The film was directed by Academy Award–nominated filmmaker Liz Garbus, who produced it alongside Dan Cogan and Darren Aronofsky. The project explored the resilience of memory through reminiscence therapy and sensory cues.
The team faced challenges in recreating memories without sensory cues, leading to a two-part technical approach that preserved emotional truth. Ethelle became an active co-creator, correcting details like the curve of a staircase or the shape of a shoe heel.
Source: googleai
