Emu Video and Emu Edit: Meta-first AI models for video and images

Meta on Thursday unveiled its two new AI tools, Emu Video and Emu Edit, in a first real-world look at the technology announced at Meta Connect in September. Emu Video is a tool that allows users to create videos from plain text prompts, while Emu Edit introduces a different approach to image editing known as Insping.

The introduction of Emu Video and Emu Edit is a strategic move for Meta, which still fits with the broader vision of the Metaverse. The company says these tools offer new creative capabilities designed to appeal to a wide range of users, from professional content creators to those looking for new ways to express ideas.

Emu Video in particular demonstrates the company's commitment to AI-driven content generation—and could be a major competitor to established names like Runway and Pika Labs, which have dominated the space until now.

Table of Contents

Toggle

Emu Video: Text-to-Video Creation

Emu Video adopts a two-step process to create videos from text prompts. First, it generates an image based on the text entered, then it creates a video from the text and the generated image. This approach simplifies the video generation process by eliminating the complex and multi-model methods used to power Meta's previous Make-A-Video tool.

The videos created by Emu Video are limited to a resolution of 512×512 pixels, but they show amazing correlation with the text questions presented. Accurately converting text into visual narratives sets Emu Video apart from most models and business solutions.

Although the models themselves are not publicly available, users can experiment with a set of predefined queries, and the results are very smooth, with minimal differences between frames.

Emu Edit: Image editing with coloring

Alongside the Emu video, Meta showcased the capabilities of Emu Edit, an AI-driven tool designed to perform various image editing tasks based on AI interpretation of natural language instructions. Emu Edit allows users to edit images with a high degree of precision and flexibility.

“Emu edit [is] A multitasking image editing model that puts state-of-the-art effects on instruction-based image editing,” says Meta's research paper, highlighting the tool's ability to accurately execute complex editing instructions.

Emu Edit's accuracy is improved by using the advanced AI technology known by Stable Diffusion. This approach ensures that edits maintain the visual integrity of the original images.

Meta's focus on developing AI tools such as Emu Video and Emu Edit includes the strategy of creating critical technologies to create the Metaverse. This includes the development of Meta AI, a personal assistant powered by the LLMA-2 large language model, and the introduction of multimodality in AR devices.

Edited by Ryan Ozawa.