Skip to Content

Lowering Image Sensor Reconfiguration Latency on Android

Mobile systems would benefit from the ability to make a tradeoff between image resolution and power consumption when doing vision tasks. While less resolution lowers task accuracy it also saves system power. Currently, resolution changes are a slow process taking over 200ms while also dropping frames. However, researchers propose a system to lower that latency

Adaptive Sampling

Modern smartphone cameras capture high-resolution images, containing millions of pixels. However, computer vision tasks operate effectively on images of much lower resolution. The research team seeks to implement an adaptive sampling block within the imaging pipeline on an FPGA. The adaptive sampling block can utilize key information from the prior image to remove unimportant pixels

LensCap

Augmented reality computing environments further the richness of the mobile experience by providing applications a continuous vision experience, where visual information continuously provides context for applications. In a modern mobile system, the typical user is exposed to potential mass aggregation of sensitive information, posing both privacy and security deficiencies.To address such deficiencies, the team proposes

LED Based Camera Pose Estimation

Augmented reality (AR) enhances the user view of the environment by overlaying the virtual information on the physical world, thereby creating an illusion to the user. For creating a good illusion, we need to accurately register the virtual world on top of real world. In fact, many applications such as AR-based classroom education demand accurate

Inside-Out Tracking with Redirected Walking for Free World Mobile Virtual Reality Navigation

Virtual Reality needs a cost-efficient, user-friendly navigation, thus the research objective is to use redirected walking for user locomotion. Research has shown that greedy algorithms can virtually alter user locomotion in a confined, physical space. Furthermore, advancements with Augmented Reality frameworks and inside-out tracking provide the opportunity for motion tracking users. Redirected Walking can prevent

Spatiotemporal Framework for Dynamic Merged Reality Content Creation

Merged Reality produces immersive user experiences by combining the real world with the virtual world to create new environments and visualizations. The objective of this research focused on building a spatiotemporal framework that assists in dynamic merged reality content creation. The tool will allow users to take computer-generated holographic data and map it to physical

Using Prefetching Algorithms to Seamlessly Download Data

The objective is to create a data distribution system that uses location data from the user’s mobile device to anticipate what information their device will need in the immediate future. The local app will then download the information so that the user can enjoy a seamless entertainment experience. In the golden age of user experience,

Real-time Illumination Estimation for Mobile Augmented Reality

High quality rendering on mobile devices is constrained by lack of resources in terms of processing power and memory leading to low visual fidelity of the rendered scenes. Previous works, try to overcome this computational challenge by offloading tasks to a powerful server. This research introduces the idea of collaborative rendering, which entails distributing the