|
|
Abstract:
Real-Time detection is a key component in various areas of computer vision, e.g. in industrial inspection, augmented reality and robotics. For many applications it has to work robustly and in real-time in order to be fully operational.
While real-time detection of well textured objects has already reached a high level of maturity, its application on low-textured or texture-less objects is still an open issue. Unfortunately, those kinds of objects play an important role in man-made environments which makes it necessary to deal with them in an efficient manner. In this thesis, we therefore present four novel methods for the efficient and reliable detection of low-textured and texture-less objects which are superior to current state-of-the-art approaches w.r.t. robustness and speed.
WebEventForm | |
---|---|
Title: | PhD? Defense of Stefan Hinterstoisser about Real-Time Detection of Low-Textured and Texture-less Objects |
Date: | 14 March 2012 |
Location: | Room 00.05.035, FMI-Building, Garching, Boltzmannstr. 3 |
Abstract: | Real-Time detection is a key component in various areas of computer vision, e.g. in industrial inspection, augmented reality and robotics. For many applications it has to work robustly and in real-time in order to be fully operational. While real-time detection of well textured objects has already reached a high level of maturity, its application on low-textured or texture-less objects is still an open issue. Unfortunately, those kinds of objects play an important role in man-made environments which makes it necessary to deal with them in an efficient manner. In this thesis, we therefore present four novel methods for the efficient and reliable detection of low-textured and texture-less objects which are superior to current state-of-the-art approaches w.r.t. robustness and speed. |
Imageurl: | |
Type: | News |
Videourl: | |
Conferencelink: |