🔥 Meta's Latest Reveal: Advanced Generative AI

posted  17 Nov 2023
Photo - Meta's Latest Reveal: Advanced Generative AI
Meta has recently introduced two artificial intelligence models, Emu Video and Emu Edit. Emu Video specializes in generating brief 4-second videos from textual descriptions or initial images, while Emu Edit focuses on detailed image editing capabilities.

Meta explains that Emu Video uses a bifurcated approach, initially generating images from text, followed by stitching these into a seamless video, thus streamlining the creative process.

Emu Edit, with its capabilities to modify backgrounds, change object colors, and introduce new elements, was developed using a specially created dataset of 10 million synthesized images.

"Unlike many generative AI models today, Emu Edit precisely follows instructions, ensuring that pixels in the input image unrelated to the instructions remain untouched," the company stated.

These new models are envisioned by developers as versatile tools for creativity, useful for artists, animators, and everyday users alike. However, at this stage, they represent an exploration of the possibilities in machine learning.