Blur2Vid Model Converts Blurry Photos into Coherent Videos

Researchers from the University of Toronto and York University introduced Blur2Vid, a model accepted by SIGGRAPH Asia 2025. By post-training a video model, Blur2Vid extracts motion information from a single motion-blurred image to generate coherent videos.

2026-08-10 ~ 2026-08-11 · 2 related posts

• Blur2Vid: Generating Videos from Motion-Blurred Images — cindy_x_wu · 2026-08-10

1 near-duplicate retellings: jon_barron