|
|
Klank, Uli
Researcher
|
Table of ContentsResearch TopicsCurrent Research Projects Teaching Inf. IX(Current Term) Students Publications |
Research Topics
- Model-Based Image Interpretation
- Gaze Tracking, Focus of Interest
- Color Detection, Adaptive Skin Color Classification
- Automatic Model Inference
- Descriptor Based 3D Localization
- Hybrid Approach 3D Localization
- Multimodal 3D Localization
- Object Tracking for Manipulation Tasks
- Hand Tracking in Combination With Object Localization
- Visual SLAM
Current Research Projects
- CoP : The project aims at the unification of vision-based sensing in the CoP (Cognitive Perception) learning and planning system. On the one hand, CoP manages the interpretation of different kinds of sensors and on the other hand it automatically acquires and maintains the knowledge about the world and objects in the world. CoP selects sensors and sensor interpretation algorithms based on their expected utility. To this end, CoP learns and improves intersensor and inter-algorithmic models for method seleciton from experience. Especially the vision system, the major sensor we use, provides several automatic model improving techniques. Improved models accelerate the perception process and provide more robust results.
Teaching General
Auf Wunsch können auch andere als die hier aufgeführten Arbeiten vorgeschlagen werden, solange sie relevante Themenbereiche abdecken. Bei Interesse an einem der unten angegebenen Projekte, würde ich mich über Nachfragen freuen.
Zu vergeben
Zu vergeben:
DA/BA/MA: Action analysis in gaze controlled videos
DA/BA/MA: Object recognition using a web based product database
DA/BA/MA: Robust recognition of 3d-shape models infered from google sketchup
DA/BA/MA: GPU optimized implementation of a descriptor based matching approach in 3D
Projekt Systementwicklung/SEP/Hiwi: Entwurf und Implementierung einer wahrscheinlichkeitsgesteuerten Exploration von Räumen mit Hilfe einer bewegten Kamera
DA/BA/MA: Define and implement a learning heuristic for a task dependant camera selection.
SEP/Hiwi: Create recognizable 3d signatures for kitchen utilities
SEP/Hiwi: Use poitning gesture recognition system for interaction with the kitchen robot
Vergeben
SEP: Beschleunigung von Objekterkennung mit Hilfe schneller Lokalisierung von
Ellipsen
MA: Action analysis in gaze controlled videos
Hiwi: Shape model inference from google sketchup
DA: Plan based optimization of computer vision tasks
BA: GPU optimized implementation of a descriptor based matching approach
Zeeshan Zia: Shape model inference from google sketchup
Rok Tavcar: Plan based optimization of computer vision tasks
Hans Dichtl: 2D Ellipse Detection for point cloud segmentation
Sebastian Marsch: Marker based navigation of smallscale robot systems in an industrial application
Florian Ferstl: GPU optimized implementation of a descriptor based matching approach
Selected Publications
-
Acquisition of a Dense 3ED Model Database for Robotic Vision, 2009, International Conference on Advanced Robotics (ICAR),(BibTeX)
-
Automatic feature generation in endoscopic images, 2008, International Journal of Computer Assisted Radiology and Surgery,(BibTeX)
