Make-a-Video release showcases Meta’s artificial intelligence generator
Type a text and AI will make a video for you
Despite recent attention towards AI text-to-image generators, researchers are now focused on AI text-to-video generators.
The Make-A-Video solution was unveiled by a group of machine learning experts from Facebook’s parent company Meta. The AI model enables users to enter a cursory description of a sight and produce a short movie from the text.
At the moment with just a few lines the AI makes movies with distorted animation and blurry subjects, but they mark a significant advancement in the field of AI content creation.
The videos are shorter than 5 seconds and run across a wide range of prompts that have no audio. Comparatively, AI image generators have progressed from producing nearly incomprehensible images to photorealistic material in just a few short years.
The near-limitless intricacy of the subject matter may also slow down the progress in video, but the reward of seamless video output will spur numerous institutions and businesses to invest significantly.
Meta says the video creation tools may be beneficial for creators and artists. However, similar text-to-image models reveal an unsettling number of possibilities, including abusive content.
