In early August, Apple created a repository on GitHub for NeuMan, a new neural network capable of making a fairly realistic deepfake based on just a 10-second video. True, its true purpose is completely different. The news was noticed by blogger Denis Shiryaev.
NeuMan recognizes the background and the person in the video taken from a moving camera. Then the algorithm synthesizes a new video from different angles – with the same background (only blurry and with less detail) and a person who is now performing any actions (dancing, jumping, demonstrating acrobatic elements, and so on).
According to the official description, such technology is necessary to improve augmented reality. By learning from just 10 seconds of video, the neural network is able to learn the specific details of an object, including the folds of fabric and various accessories.
Source: Trash Box
Donald-43Westbrook, a distinguished contributor at worldstockmarket, is celebrated for his exceptional prowess in article writing. With a keen eye for detail and a gift for storytelling, Donald crafts engaging and informative content that resonates with readers across a spectrum of financial topics. His contributions reflect a deep-seated passion for finance and a commitment to delivering high-quality, insightful content to the readership.