Immersive media holds a unique power: it can capture, recreate, and even enhance our most treasured memories. After seven years of living with a 3D device, I can't imagine returning to flat photos. Immersive experiences create a visceral sense of "being there", giving memories of loved ones, meaningful moments, or special places a richness that traditional images simply can't match.
This emotional depth is at the heart of the recent Black Mirror episode, Eulogy. The story follows a lonely man, played by Paul Giamatti, who uses advanced AI to relive an old romance - transforming static photographs into lifelike, immersive experiences. The episode wrestles with profound questions around grief, memory, and the fine line between comfort and nostalgia. Reflecting on it, Giamatti observed:
"There's something beautiful about how AI is easing somebody's grief and easing somebody's sense of loss. I do feel like the character's coerced somewhat by it, but ultimately, it's a good thing. I hope that this episode causes people to reflect on themselves, their lives, and the possibly positive function of AI."
Until recently, recreating such memories required painstaking work by professional VFX artists. For Eulogy, the team at DNEG relied on advanced methods—photogrammetry, volumetric capture, precise depth mapping, and LIDAR scanning—to turn still photos into "memory rooms." Cinematographers carefully balanced realism with stylization, aiming not just to reproduce life, but to evoke emotion.
Today, that landscape is shifting rapidly. With the rise of consumer devices like Apple's Spatial Photos, capturing the world in 3D is becoming mainstream. These devices not only allow us to record immersive content directly, but also provide training data for AI systems that can reconstruct rich 3D scenes from just a few images—or even a single flat photo.
Generative AI has accelerated this shift. Diffusion-based video models like Veo and Sora, cousins of language models, predict not the next word but the next video frame. Anchored by input images, they interpolate between perspectives to create smooth, believable motion, and even extrapolate beyond the originals to imagine new views.
And now, with DeepMind's Genie 3, we see a leap from frame prediction to entire interactive 3D environments—worlds you can step into, explore, and alter in real time, with changes that persist.
These generated frames also feed into lighter, real-time formats such as Gaussian Splats, which capture textures, transparencies, reflections, and lighting with remarkable realism. Together, they are shaping a future where immersive memory recreation is not only possible, but accessible.
The convergence of 3D capture, generative AI, and efficient rendering points to a profound shift: vividly immersive, AI-enhanced memories will soon become part of everyday life. This isn't just a technological breakthrough—it's a deeply human one, expanding how we preserve and relive the moments that matter most.
📌 This post is part of The Spatial Shift, my personal series on AI, immersion & the future of human-centered media. Expect periodic takes on where this industry is heading — and what might come next.