Details

Type

  • Bachelor Thesis
  • Student Project
  • Master Thesis

Persons

1-3
Workgroup
Rendering and Modeling

Motivation

In fast-paced movements such as sports (skiing, running, cycling, diving, among others) and emergency operations (fire, alpine, underwater, or earthquake rescue), especially in poor sight due to weather conditions or visual pollution, it is very useful to crucial to display important information about the environment geometry, e.g., slope, obstacles, distances, optimal navigation paths. This can be done as Augmented Reality on glasses in the form of enhanced contours to highlight obstacles, illustrate slopes or distances, navigation cues, and all of it best in a most unobtrusive way while focusing on the time-critical tasks. 

Description

This requires both quick scanning of the relevant surrounding geometry and acquiring video for registration with known environment models, as well as processing all of this in real-time, which is an extra challenge to implement on mobile hardware with little processing power. 

Tasks (depending on PR/BA/DA and number of students)

  • Connect a LIDAR scanner (e.g., https://www.livoxtech.com/mid-360) with an Ethernet2USB-C connector to a smartphone to enable mobile acquisition of scans, with a data loader
  • Connect a helmet camera (Insta360 or GoPro 360) to a smartphone to transfer images in real-time
  • Localize yourself in the 3D environment (e.g., terrain data) using GPS, IMU data, and registering with the acquired geometry
  • Localize yourself in the environment using the acquired video data, using existing geolocated photos
  • Unobtrusive AR display of information in 2D (e.g., contour edges, numerical information)
  • Unobtrusive AR display of information in 3D (e.g., slope, distances, navigation cues)
  • Evaluation of geometry and photo registration quality and runtime, user studies of AR displays

Requirements

  • Knowledge of English (source code comments and final report have to be in English)
  • Knowledge of basic geometry processing is required, experience in mobile coding (Fluttr/Kotlin) is a plus since all algorithms have to run on a smartphone
  • Going snowboarding or skiing in the 2025 winter season with scanning equipment is also a plus

Environment

The project should be implemented platform independent (Android, iOS).

A bonus of €500/€1000 if completed to satisfaction within an agreed time-frame of 6/12 months (PR/BA or DA)

Responsible

For more information please contact Stefan Ohrhallinger.