Abstract:
The proposed system targets a mobile phone android application using Augmented Reality (AR) as means of user interface. AR is a concept of integrating virtual 3-dimensional objects on 2-dimensioanl user screen creating an illusion of a 3-dimensional, real time user experience of an inherently simulated object in the real environment. The system will serve as a medium for extensive class of students focusing on assisting them to build better semantics for biological concepts through E-learning. The application focuses on four main human organs (heart, brain, stomach and lungs). The user will be able to view the 3D model of the four mentioned organs augmented on their books, saving not only learning time but also to provide a unique perspective to the curious nature of the human mind to understand the inherited complexity of its constituent anatomy. The proposed system achieves this by providing unique interfaces for the user to understand the working functionalities of human anatomy by simulating the 3D models. Converging to the recursively natural approach in machine learning we choose, state of the art, pre-trained convolutional neural network (CNN) as system’s classifier to recognize particular human organs such as Brain, Heart Lungs and Stomach from input camera frames. Once recognized, the particular target model is loaded from the model repository and simulated on user screen for interaction. To ensure maximal student access, the system is deployed as an android application, fine-tuned with respect to the execution parameters such as CNN and anatomy model’s storage size kept to the minimum. For modelling and animation of organs, 3D modelling software like Blender, Cinema 4D and Autodesk Maya is used respectively. Unity platform is used for integration of different components and to develop Augmented Reality based mobile application.
