Video creation processes are becoming more dynamic as platforms introduce new tools for digital creativity. Animations rendered on-demand are becoming more closely aligned with what the future of entertainment and creation in the metaverse will entail.
Meta announced an AI-video generator in September that renders video from text. Make-A-Video can create a video on demand using any descriptive text or existing image prompts, according to the tech giant. The technology uses “paired text-image data,” Meta describes in a press release, and the company aims to make the tech publicly available in the future with the goal of “pushing creative expression forward by giving people tools to quickly and easily create new content.”
Just a few days later, Google released its own AI system to generate videos from text prompts. Imagen Video was programmed on 14 million video-text pairs and 60 million image-text pairs according to Google, and can conceptualize depth in 3D animated movement. In a paper, Google describes the program as “a high definition text-to-video model” with “classifier-free guidance for fast, high quality sampling,” offering creators “the ability to generate diverse videos and text animations in various artistic styles and with 3D object understanding.”
Microsoft is bringing its existing DALL-E 2 services to the public with first-party apps. This AI-powered system currently renders images from text and is being used by brands ranging from Nestlé and Stitch Fix to architectural firms for product and building conceptualization. In mid-October, Microsoft announced its plan to integrate DALL-E 2 with its recently announced Designer app and Image Creator tools: now available in Bing and Microsoft Edge. According to OpenAI, more than 1.5 million users are creating over 2 million images a day using DALL-E 2. The two companies told TechCrunch that they plan to “continue to explore solutions that harness the power of AI and advanced natural language generation.”
These tools indicate a future in which creators can animate and visually create interactive film, video and art with ease and flexibility. Creating content dynamically, and rendering it spontaneously using text, opens up a breadth of possibilities for the future of visual creation and creativity and may make it accessible to a wider user base.