Academy of Interdisciplinary Studies Division of Integrative Systems and Design 246 Division of Integrative Systems and Design Augmented Reality Metaverse Supervisor: Tristan Camille BRAUD / ISD Student: BOW Wing Yin / COMP Course: UROP 1100, Fall This project is dedicated to the advancement of augmented reality technologies with a focus on the development of an immersive metaverse. Key areas of exploration include place and object recognition, citywide AR applications, AR-enhanced workspaces, and web browsing integration, as well as AR content creation and the utilization of non-fungible tokens (NFTs). The primary objective is to pave the way for a comprehensive VR guide tour experience. Presently, the project is concentrated on refining techniques for the retrieval and integration of 3D objects within the AR camera environment. Through these efforts, the project aspires to contribute significantly to the evolution of augmented reality and its applications in creating interactive and engaging virtual experiences. Augmented Reality Metaverse Supervisor: Tristan Camille BRAUD / ISD Student: CHAN Shu Pui / COSC Course: UROP 3100, Fall Accurate and robust real-time visual localization is crucial for various applications, including autonomous driving, Augmented Reality (AR), and eXtended Reality (XR). It enables the establishment of a dead anchor for Inertial Measurement Unit (IMU) tracking and allows multiple users to share the same coordinates in virtual space. Hierarchical Localization is a state-of-the-art visual localization method that globally retrieves location hypotheses and matches local features to achieve a precise and robust 6-DoF positioning. This report describes the modifications made to hloc, a Python library that implements Hierarchical Localization, aimed at reducing its runtime and managing the map to allows for expansion and update. Generative AI in Interactive Multimedia Applications Supervisor: Tristan Camille BRAUD / ISD Student: LAM Ching Yin / CPEG Course: UROP 1100, Fall With the relatively recent release of generative AI tools like large language models, the limit of what content computers can provide has been pushed a lot further. For instance, chatbots powered by large language models can respond to simple prompts with useful output, and image generation models provide images based on text prompts as well. There is massive potential behind this technology that can be used in multimedia applications, such as in 3D environments like gaming, but these have not been explored deeply yet. This project explores how generative AI can be used to generate content for real-time gaming, with concept demonstrations conducted in the game engine Unity.
RkJQdWJsaXNoZXIy NDk5Njg=