[Experiment] I trained a model on childhood photos to simulate memory recall

A developer fine-tuned the SDXL model using 60 childhood photographs from a limited family archive to simulate memory recall. Instead of faithfully reconstructing the images, the model generated unstable variations, producing familiar yet non-existent spaces, faces, and fragments. This experiment aimed to revisit a period of life through AI-generated interpretations of personal memories. More project details and tutorials are available via YouTube, Instagram, Patreon, and Uisato Studio.