Martin Feick

Doctoral Researcher

Loading...

Research Interests

Virtual RealityMixed realityFabricationTangible User Interfaces3D User InterfacesRobotics
 

Open Thesis Topics

In this thesis you will investigate virtual object deformations through the use of tangible proxy objects in VR. This requires the implementation of a deformable physical prototype (e.g., bendable, twistable, etc.) which acts as a counterpart for several virtual objects. The goal of this thesis is the creation of tangible proxies providing realistic haptic sensations in VR validated through a user study.

 

Focus

This work will mainly focus on implementing a prototype. Thus, solid programming (C#, Unity3D) and/or hardware skills (Arduino) are necessary. In addition, a user study may be carried out to evaluate the prototype.

 

Prerequisites

  • Read the following papers [1] [2] [3] (might require VPN)
  • Be familiar or get familiar with the concept of tangible interfaces [4]
  • Background or interest in Virtual Reality
  • Experience with Unity3D, 3D printing and/or Arduino
  • Completed HCI and/or Physical Computing lecture and ideally already attended at least one seminar at our chair

 

How to apply

Please send me an email with the following pieces of information (if you do not answer every point, your application will not be considered):

  • When you plan to start the thesis
  • When you plan to finish the thesis
  • A short motivational statement (max 0.5 pages) why this topic is interesting for you
  • Your transcript of records and your CV

Human as well as robotic grasp prediction has experienced the development of numerous methods for learning suitable grasping motions with respect to a wide variety of objects.In the field of robotics, various (learning-based) approaches have been proposed [1][2]. An enormous amount of methods have been applied for predicting a human's grasping motions as well. However, the majority of these methods utilise subsymbolic learning models that do not allow more detailed insights regarding the decision making of the system. This thesis aims to investigate the possibilities of applying hybrid artificial intelligence approaches to the field of human grasp prediction (explainable AI).

 

Focus

The student is expected to work on the conceptualization of hybrid learning approaches for making explainable predictions in the field of human grasp prediction. Optimally, the student implements a prototype (Python) for demonstrating the approache's applicability.

 

Prerequisites

  • The following thesis provides a gentle introduction to deep-learning based grasp prediction, familiarize yourself with it's basic ideas [3]
  • Also, the following TechTalks article provides a reasonable and easy to grasp introduction to the concept and potential importance of hybrid AI [4]
  • You should have a good command of a variety of machine- and deep-learning algorithms
  • Be familar with implementing basic learning applications in Python using Scikit-learn and/or Keras
  • Last but not least, be enthusiastic about the investigation of potential research directions in a widely unexplored field!

 

How to apply

Please send an application to Niko Kleer or Martin Feick containing the following information. (if you do not answer every point, your application will not be considered):

  • When you plan to start the thesis
  • When you plan to finish the thesis
  • A short motivational statement (max 0.5 pages) why this topic is interesting for you
  • Your transcript of records and your CV
Teaching a robot basic motions for grasping objects is a complex task that research has been tackling from numerous perspectives. Common approaches for deriving basic grasping motions include applying Deep Learning to large collections of images [1]. In other cases, additional hardware such as a glove might be used for capturing a human's motions which can be translated to a more specific grasping type [2]. Unfortunately, many of these approaches cannot be applied in real world applications. This thesis aims to investigate the possibility of teaching a robot basic grasping motions based on raw data while refraining from the usage of additional, and at the same time, cumbering hardware.

 

Focus

The student is expected to work on the implementation of a system that allows predicting the human's grasping motion while investigating the applicability/potential of various learning algorithms.

 

Prerequisites

The following thesis provides a gentle introduction to deep-learning based grasp prediction, familiarize yourself with it's basic ideas [3]:

  • You should be familiar with the concept of Deep Learning and how to implement CNNs using Keras
  • Previous exposure to OpenCV and applying learning algorithms to videos is a plus
  • Last but not least, be enthusiastic about the investigation and application of learning algorithms in the context of robotics!

 

How to apply

Please send an application to Niko Kleer or Martin Feick containing the following information. (if you do not answer every point, your application will not be considered):

  • When you plan to start the thesis
  • When you plan to finish the thesis
  • A short motivational statement (max 0.5 pages) why this topic is interesting for you
  • Brief summary on your prior knowledge regarding learning algorithms (specifically reinforcement learning) 
  • Your transcript of records and your CV

 

 

Open B.Sc. / M.Sc. Thesis Topics

If you are interested in doing a Bachelor's or Master's thesis in my area of interest, feel free to contact me. Please check the open topics at the top of this page beforehand (if there are any).

 

Projects

CAMELOT

TRACTAT

 

 

Academic Services

Reviewer: CSCW 2021 Full papers, CHI 2021 Full Papers, CSCW 2020 Full papers, CHI 2020 Late-Breaking-Reports, INTERACT 2019 Full papers, DIS 2019 Work-In-Progress, CHI 2019 Late-Breaking-Reports, ISS 2018 Posters, SUI 2018 Full papers, CSCW 2018 Posters, HRI 2018 Late-Breaking-Reports

PC: TEI'21 WiP


Publications ()

Loading...