Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revisionPrevious revision
Next revision
Previous revision
teaching:gsoc2017 [2017/03/05 19:09] – [Topic 2: Realistic Grasping using Unreal Engine] ahaiduteaching:gsoc2017 [2017/03/17 18:21] (current) – [Topic 1: Multi-modal Cluttered Scene Analysis in Knowledge Intensive Scenarios] balintbe
Line 95: Line 95:
 would start from (but not necessarly limit ourselves to) the implemen- would start from (but not necessarly limit ourselves to) the implemen-
 tation of two state-of-the-art algorithms described in recent papers: tation of two state-of-the-art algorithms described in recent papers:
-[1] Ilya Lysenkov, Victor Eruhimov, and Gary BradskiRecognition + 
-and Pose Estimation of Rigid Transparent Objects with a Kinect Sen- +[1] Aleksandrs EcinsCornelia Fermuller and Yiannis AloimonosCluttered Scene Segmentation Using the Symmetry ConstraintInternational Conference on Robotics and Automation(ICRA2016
-sor2013 Robotics: Science and Systems Conference (RSS), 2013.+
 [2] Richtsfeld A., M ̈ [2] Richtsfeld A., M ̈
 orwald T., Prankl J., Zillich M. and Vincze orwald T., Prankl J., Zillich M. and Vincze
Line 106: Line 105:
 **Task Difficulty:** The task is considered to be challenging, as it is still a hot research topic where general solutions do not exist. **Task Difficulty:** The task is considered to be challenging, as it is still a hot research topic where general solutions do not exist.
      
-**Requirements:** Good programming skills in C++ and basic knowl- +**Requirements:** Good programming skills in C++ and basic knowledge of CMake and ROS. Experience with PCL, OpenCV is prefered.
-edge of CMake. Experience with PCL, OpenCV is prefered.+
  
 **Expected Results:** Currently the RoboSherlock framework lacks good perception algorithms that can generate object-hypotheses in challenging scenarios(clutter and/or occlusion). The expected results are several software components based on recent advances in cluttered scene analysis that are able to successfully recognized objects in the scenarios mentioned in the objectives, or a subset of these. **Expected Results:** Currently the RoboSherlock framework lacks good perception algorithms that can generate object-hypotheses in challenging scenarios(clutter and/or occlusion). The expected results are several software components based on recent advances in cluttered scene analysis that are able to successfully recognized objects in the scenarios mentioned in the objectives, or a subset of these.
Line 150: Line 148:
 robotics. robotics.
  
-**Expected Results** We expect to enhance our currently developed robot learning game with realistic human-like grasping capabilities. These would allow users to interact more realistically with the given virtual environmentHaving the possibility to manipulate objects of various shapes and sizes will allow to increase the repertoire of the executed tasks in the game. Being able to switch between specific grasps will allow us to learn grasping models specific to each manipulated object.+**Expected Results** We expect to be able to load URDF models of 
 +varios robots (e.g.  PR2) and be able to control them through ROS 
 +in the game engine In a similar fashion to a robotic simulator.
  
 Contact: [[team/andrei_haidu|Andrei Haidu]] Contact: [[team/andrei_haidu|Andrei Haidu]]




Prof. Dr. hc. Michael Beetz PhD
Head of Institute

Contact via
Andrea Cowley
assistant to Prof. Beetz
ai-office@cs.uni-bremen.de

Discover our VRB for innovative and interactive research


Memberships and associations:


Social Media: