We are looking for interested students for BA or MA theses, for being a Hiwi oder for helping us in our research areas. Please email Prof. Deussen and make an appointment.
For the winter term 2022/23 there will be a presentation of current projects on Thu, the 27.10 at13:30, during the seminar slot.
Cluster of Excellence Collective Behaviour
Our group is part of the new excellence cluster of the University of Konstanz "Centre for the Advanced Study of Collective Behaviour" (https://www.exc.uni-konstanz.de/collective-behaviour/), a cross-disciplinary effort for the study of collective behaviour between the University of Konstanz and the Max Planck Institute of Animal Biology.
Here we are offering Bachelor and Master projects/thesis (as well as HiWi jobs) in the following areas:
- 3D reconstruction and object classification with a focus on underwater ecosystems
- Simplification of trajectories from motion tracking
- Using machine learning to track identities in trajectories
This is not an exhaustive list but instead should give you a general idea of the topics within this cluster. If you are interested to work in any of these topics in a cross-disciplinary collaborative setting, please feel free to write an email to oliver.deussen@uni-konstanz.de
Visualization: Color selection for Euler Diagrams
Bachelor / Master
Euler Diagrams can be created automatically, but assigning colors to sets is only little researched.
For this project you will:
- Improve upon the random assignment of colors for the individual set representations
If you are interested in Euler diagram visualizations,
please contact: Patrick Paetzold
Visualization: spEuler integration of datapoints
Bachelor / Master
We can use Euler diagrams to visualize complex relationships between sets. One such technique to visualize these Euler diagrams is spEuler.
However, it is also interesting to visualize individual datapoints, such as images or text directly in the Euler diagram.
For this project you will:
- Fill the shape of Euler diagrams with individual objects, such as words or images
- Use preexisting algorithms, such as Wordle or ShapeWordle
If you are interested in Euler diagram visualizations,
please contact: Rebecca Kehlbeck
Visualization: Frontiers of hallucination
Bachelor / Master
For this project you will:
- Build interface to investigate the border cases of hallucinating objects using Dall-E.
- Mask parts of an image and insert individual objects via text prompt (inpaiting) and compare the results across different domains
If you are interested,
please contact: Rebecca Kehlbeck
Computer Graphics: Procedural distribution of vegetation
Bachelor / Master
For creating realistic landscapes, it is necessary to model natural vegetation distributions correctly.
For this project you will:
- Determining planting patterns in different landscape types
- Identify factors that influence these patterns
- Define rules for those patterns
If you are interested,
please contact: Till Niese
Computer Graphics: LoD - Abstraction of large ecosystems
Bachelor / Master
Interactively rendering large ecosystems is not trivial and offers different research areas.
For this project you will:
- Representing the ecosystem not through geometry but by using limited metadata describing the individual plants
- Create spatial clusters based on these metadata
- The generation of suitable representations for the identified clusters
If you are interested,
please contact: Till Niese
Visualization: Reduce uncertainty in parallel coordinates
Bachelor / Master
The Gestalt principle of continuity can lead to false perceptions by users, it is also prevalent in parallel coordinates.
For this project you will:
- Remove as much of this uncertainty as possible
- Use a variety of approaches such as pattern recognition, interaction, etc. to produce low uncertainty visualizations
If you are interested,
please contact: Yumeng Xue
Visualization: Interactive exploration over large-scale time series data
Bachelor / Master
Large-scale time series data poses challenges to user interaction exploration, such as the real-time availability of interactions, and the degree of freedom of interaction.
For this project you will:
- Implement queries on the time series by free-sketching, find a subset that matches the sketch, and apply the queried data to the rendering enhancement
- For example if the user queries for line bundles that are not present in the initial rendering (low density, but still exist in the data), you should improve the rendering results by presenting this trend in the graph
- Your query should be as real-time as possible
If you are interested,
please contact: Yumeng Xue
e-David Painting Robot
Various Projects available
The e-David painting robot is under ongoing development and both its works and the machine are frequently exhibited in prominent locations. In order to further the development of the machine several topics can be addressed in a Bachelor's or Master's thesis:
- Implementation of new painting processes, based on our feedback mechanism.
- Robotic brush and paint handling for optimizing tool use during the painting process.
- Hardware and driver development for new robots, to allow their use as e-David machines.
- Exploring the applicability of machine learning techniques to generate painting commands for the robot.
Some of these projects can involve working with hardware and electronics. Others are mostly about graphics and image processing. The focus of your thesis can be set based upon your interests in these topics.
If you are interested in art, robotics and the real world implementation of computer graphics techniques, please contact Marvin Gülzow.
Planned Projects:
- Painting Simulation and Visualization: You will implement a C++ module for the e-David control software, which allows system users to preview robot movements for different use cases. A technical view should display movements in 3D, so newly developed stroke techniques can be analyzed for issues like collisions. An artistic view should show a simulated painting, based on a set of brush commands.
- Generative painting: You will implement a generative model which can produce variations of an object, like a tree, a flower or some other category. The model should output both a visualization as well as strokes, which make use of brush dynamics in order to represent the object. It should be possible to parametrize the size, rotation and other parameters, to compose an entire painting of several generated objects.
- Automatic Brush Calibration: Despite the robot being a precise machine, brush deformation causes significant errors in stroke placement. Two potential methods exist for compensating this: First, you could consturct a device for optical inspection of the brush, where the robot presents the current brush and your device measures the tool tip offset. Second, you could implement a method which paints certain calibration patterns, which are recorded by the visual system and are then used to compute a prediction model for brush behaviour at varying angles and pressure levels.
- Automatic Color Mixing: In order to realize paintings, a human artist mixes available colors to obtain a desired new tone. In this project you will use visual feedback to measure provided colors, construct a palette system in which to mix pigments automatically and finally implement a method which allows the robot to predict mixing results or to experimentally mix until a desired tone has been achieved.
