Reconstructing Videos In Just 14 Seconds: Meta AI’s Fairy Accelerates Video Synthesis by 44×

A Meta GenAI research team introduces Fairy, a versatile and efficient video-to-video synthesis framework. Fairy stands out for its ability to generate high-quality videos at remarkable speed, producing 120-frame 512×384 videos in just 14 seconds, surpassing previous works by a factor of at least 44×.

ai artificial intelligence computer vision & graphics deep-neural-networks diffusion model framework genai generate least machine learning machine learning & data science meta meta ai meta genai ml quality research research team speed synthesis team technology video videos video-synthesis video-to-video