Self-Trained Motion Model Plus Image Diffusion Creates Datamosh Animations

Creator pixlpa combined a self-trained motion model with image conditioning and a small diffusion model for inpainting occluded regions, producing trippy datamosh-style animations. He noted the improved motion conditioning ties form and motion together but can fall into feedback loops.

2026-09-14 ~ 2026-09-15 · 4 related posts

• Dev chains a motion model with an image diffusion model to make glitchy datamosh animation — pixlpa · 2026-09-14
• Dev combines custom motion model with image diffusion to make datamoshed animation — pixlpa · 2026-09-15
• Motion conditioning improvements tighten form-motion coupling in video gen, but feedback loops lurk — pixlpa · 2026-09-15

1 near-duplicate retellings: David_Rudnick