Drones are transforming surveillance practices, but how do you best manage massive volumes of video feeds? Instead of the typical 2D video wall of drone feeds, deploy a dynamic 3D model continuously updated from drone video and 3D reconstruction, which also also optimizes swarm control. May enhance military operations, disaster response, and search-and-rescue by improving efficiency, situational awareness, and scalability.

To enhance the realistic appearance of virtual environments, we project video onto a 3D model using a LIDAR-scanned point cloud while addressing complex challenges, including conflicting viewpoints, camera alignment, projection through floors, video mosaic blending, and seam correction. Each solution is designed to maintain scene accuracy, ensure aesthetic realism, and sustain real-time frame rate for practical, live applications. 

This digital twin of SkyWater Technology’s silicon wafer factory enables SkyWater to explore multiple approaches to improving manufacturing efficiency. Its data visualization dashboard presents real-time factory floor metrics, simplifying decision-making. The backend features a macro simulation of a wafer metrology facility, modeling the creation and movement of silicon wafers and feeding the data to the frontend.

The Crime Report project’s Traffic Simulation Digital Twin is a real-time, dynamically generated traffic simulation of the UCF campus. Built in the Unity 3D game engine, this system integrates open-source data from Cesium, OpenStreetMap, and the Simulation of Urban MObility platform, creating a scalable and responsive framework for traffic simulations. Designed for flexibility and extensibility, it allows application-level use cases to be developed without the need to rebuild the underlying digital twin environment. 

This embodied AI/augmented reality assistant integrates with smart home device networks. While you wear an augmented reality headset, this Assistant appears as a floating presence beside you, responding to your voice commands. Saying “Help me get ready for bed” prompts the Assistant to dim the lights, adjust room temperature, and play soothing music via a smart speaker device.

Our experimental art experience deploys augmented reality to blur the line between our digital and physical identities. The piece centers around a mural installed in our lab space, and lets you use a smartphone camera to reveal a virtual scene featuring VARLab students. It’s intended as an evolving digital yearbook and a way to connect with past and present classmates.

One goal for this project is to produce a digital twin of the Earth, including live weather, traffic, and place data, to explore Google Earth VR in a CAVE environment. EarthXR’s design allows it to be deployed to many XR systems, including alternative 3D displays, VR headsets, and mixed reality devices.

This embodied Artificial Intelligence digital twin is designed to create a sense of presence, enabling natural and intuitive interactions to accomplish various tasks. Powered by in-house large language models and advanced image recognition techniques, this AI can engage with humans dynamically, interpreting facial expressions to assess emotional states and adjusting its responses accordingly.

Our life-sized, interactive human anatomical model is presented on the VTable; its integration with a medical database provides accurate anatomical labeling. The virtual human model includes all major anatomical structures, each of which can be explored and interacted with for the ultimate immersive anatomy learning experience.

This real-time warehouse facility simulation developed with Siemens Material Handling accurately replicates Siemens machines, their operations, and how they process tasks. The configurable 3D model allows Siemens to customize facility layouts, and uses actual CAD data to ensure accurate representations of machinery within the virtual environment. The simulation analyzes and optimizes parcel throughput.