Thermodynamics-informed neural networks for physically realistic mixed reality

Computer science advances in the last decades led us to experience in a relatively short lapse of time three major technological innovations: the personal computer, the Internet, and mobile devices. Currently, we are at the beginning of a fourth paradigm of computing innovations involving immersive technologies such as Virtual Reality (VR), Augmented Reality (AR) or Mixed Reality (MR). All this is possible due to huge advances in machine learning techniques and hardware improvements applied to computer graphics and computer vision. It is clear that this new paradigm seeks to revolutionize technology in the next years and will have a great impact in society such as smart cities [1, 2], new teaching methods [3] or economic paradigms [4]. Technology companies have already created numerous digital platforms, such as the Metaverse [5, 6] or the Omniverse [7, 8], in order to develop their own immersive technologies.