DaDynamicOcclusion

Chair for Computer Aided Medical Procedures & Augmented Reality
Lehrstuhl für Informatikanwendungen in der Medizin & Augmented Reality

Dynamic Occlusion for Mobile Augmented Reality Applications

Motivation

First have look at the following video and things might get clear:


Among various effects occlusion has an essential visual impact on depth perception for human beings. The handling of occlusion in the filed of Augmented Reality (AR) is important for displaying correct 3D information such that it is immersive for the user. In AR scenarios, up to now, persons moving within the camera view that are used for showing augmentations are usually disturbing the depth perception since they are not tracked in the normal case. This holds even more for scenarios with a mobile camera for the augmentations. A static People Tracking system is capable of keeping track of persons within the observed area. With knowledge of the positions of the tracked persons mobile clients could be able to use this information to handle occlusion effects. In this thesis the position and shape information of persons from a static people tracking system, developed at the chair, should be used to analyse the camera images of a mobile camera. Depending on this information a first guess can be made about the shape of the person and the spatial relation between the shown virtual information. Further image processing should lead to a qualitativ better estimation of the person's shape within the camera view which can than be used for a correct occlusion handling. This image processing is the topic to be researched during this thesis.

Requirements

  • You should be motivated in AR and should have knowledge of basic image processing algorithms.
  • Knowledge of C++ is obligatory.
  • Knowledge of OpenCV is auxiliary.

Contacts:


ProjectForm
Title: Occlusion Handling for AR in Dynamic Camera Szenarios
Abstract:  
Student: Lorenzo Pirritano
Director: Prof. Gudrun Klinker(Ph.D.)
Supervisor: Christian Waechter
Type: DA/MA/BA
Area: Computer Vision
Status: finished
Start: 2010/08/15
Finish: 2011/02/15
Thesis (optional):  
Picture:  


Edit | Attach | Refresh | Diffs | More | Revision r1.8 - 15 Dec 2011 - 15:01 - ChristianWaechter