Scientists have created a technology that can produce 3D facial reconstruction from a 2D image, in short, a 3D selfie.
Scientists from the University of Nottingham and Kingston University developed a web application through which people can upload an image and the app will convert it to a 3D image of their face within seconds. Till now, more than 400,000 users gave tried it. Users can try experiment this new feature here: http://www.cs.nott.ac.uk/~psxasj/3dme/
They named the research as ‘Large Pose 3D Face Reconstruction from a Single Image via Direct Volumetric CNN Regression’ and was led by Aaron Jackson along with Adrian Bulat.
Researchers create 3D printed objects that change shape
The feature was made by using a Convolutional Neural Network (CNN) – an area of artificial intelligence that uses machine learning for providing computers the skill of learning, devoid of being programmed. The CNN was then able to reconstruct 3D facial geometry from a single 2D image. It can also show the parts of face that are not visible. Though not perfect, the new technology is still mesmerizing for many, reported Science Daily.
Research supervisor Georgios Tzimiropoulos expressed, “The main novelty is in the simplicity of our approach which bypasses the complex pipelines typically used by other techniques. We instead came up with the idea of training a big neural network on 80,000 faces to directly learn to output the 3D facial geometry from a single 2D image.”
Jackson further explained, “Our CNN uses just a single 2D facial image, and works for arbitrary facial poses (e.g. front or profile images) and facial expressions (e.g. smiling).”
“The method can be used to reconstruct the whole 3D facial geometry including the non-visible parts of the face,” added Bulat in their original publishing.
Apart from face and emotion recognition, this technology can also be used to personalize computer games, enhance augmented reality and also allow people to try on online accessories. It could also be beneficial in simulating plastic surgery results or understanding diseases like autism or depression.
Dr. Vasileios Argyriou exclaimed, “What’s really impressive about this technique is how it has made the process of creating a 3D facial model so simple.”