A research team from the University of North Carolina and the University of Maryland has developed an advanced facial age conversion system called "MyTimeMachine" (MyTM), which can personalized simulate facial morphology of different ages based on photos provided by users. This technology breaks through the limitations of traditional methods and achieves high-fidelity age regression and age growth effects by combining global aging models and personal selfie photos while well retaining personal identity characteristics. The emergence of MyTimeMachine has brought more realistic and personalized visual effects production solutions to the film and television special effects, advertising and other industries, and also provided users with a new way to experience the charm of technology.
Recently, a research team from the University of North Carolina and the University of Maryland launched a new technology called "MyTimeMachine" (MyTM), an advanced system that can personalize facial age conversion. As long as 50 photos are uploaded, the system can Learn your facial features and estimate what you will look like at different ages.
By combining a global aging model and the user's personal selfie photos, the technology allows users to achieve high-fidelity age regression (de-aging) and age growth (aging) effects while maintaining individual identity characteristics.
With the development of science and technology, facial aging has become a topic of great concern. Traditional facial age conversion methods often fail to accurately reflect an individual's appearance at the target age, resulting in generated images that are far from reality. The innovation of MyTimeMachine is that it uses 50 selfie photos provided by users to train a network of adapters to achieve personalized age conversion.
The system works by first generating base facial features using a global age encoder, and then combining these features with the user's personal photos through a network of adapters to generate new facial images. The research team also designed three loss functions to enhance the personalization capabilities of the adapter network and ensure that the generated images maintain the user's identity characteristics during age changes. This technology is not only suitable for still images, but can also be extended to video processing to achieve high-quality, identity-preserving and time-consistent aging or de-aging effects.
In practical applications, this personalized age conversion technology can be widely used in many fields such as film and television special effects, advertising industry, etc., making the visual effects more realistic. Users only need to upload their personal photo collections to easily experience the age transition process and feel the magical changes brought about by technology.
Project entrance: https://mytimemachine.github.io/
The emergence of MyTimeMachine technology marks a new height for personalized age conversion technology. Its application prospects in various fields are broad and its future development is worth looking forward to. The convenience and efficiency of this technology will undoubtedly change people's understanding of age conversion technology and promote technological innovation in related fields.