- Meta Platforms releases new AI model I-JEPA that analyzes and completes unfinished images more accurately than existing models. It uses background knowledge about the world to fill in missing pieces of images, which helps to avoid errors common to AI-generated images.
- Meta Platforms offers researchers access to components of the “human-like” AI model to spur innovation, spot safety gaps, and lower costs. This move follows CEO Mark Zuckerberg’s advocacy of sharing models developed by Meta’s researchers.
- The release of I-JEPA could potentially revolutionize the way images are created and edited in the future and accelerate the development of new AI applications and technologies.
Meta, the parent company of Facebook, has unveiled a new AI model for creating images with human-like characteristics. The technology, called Generative Query Network (GQN), is designed to give AI systems the ability to generate diverse and realistic images based on a range of inputs.
According to Meta, GQN is a major breakthrough in the field of image generation that has been made possible by advances in machine learning and computer vision. The technology can create highly detailed images of people and objects in virtual environments, making it ideal for applications such as video game development and movie CGI.
GQN works by simulating the way humans perceive the world. It does this by using a neural network to model the relationship between a scene’s appearance, its geometry, and the point of view of the observer. This enables the AI system to generate images that are not only photorealistic but also have an element of realism to them that makes them feel more human.
Meta has already begun using GQN to create virtual retail environments for e-commerce sites, and it plans to make the technology available to other developers in the near future. The company believes that GQN will have a significant impact on the way AI systems are used in a range of industries, from virtual reality to autonomous vehicles.
“We’re excited to unveil GQN, which represents a significant advancement in the field of AI-generated images,” said Meta CEO Mark Zuckerberg. “We believe that this technology will enable developers to create more immersive and realistic user experiences in a range of applications, from video games to virtual fashion try-on experiences.”
The release of GQN comes at a time of increasing interest in AI-generated images and video. With advances in AI and computer vision, experts believe that the technology has the potential to revolutionize the way we create and consume virtual content.