Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.
Research of the MVP Group                       http://mvp.visual-navigation.com                                          ...
Visual Localization for hybrid Environmental ModelingMachine Vision and Perception Group @ TUM                            ...
MVP   Machine Vision and Perception Group @ TUM                                                  Example of a hybrid model...
Result of a visual localization                                             6DoF pose of the robot is calculated fom singl...
Reconstruction of a hybrid (appearance/geometry) modelMachine Vision and Perception Group @ TUMMVP
Visual HomingMachine Vision and Perception Group @ TUM                                            The system is capable of...
Biologically Motivated Navigation                                                      Obstacle avoidanceMachine Vision an...
MVP   Machine Vision and Perception Group @ TUM                                                  Work on Optimal Sensor Mo...
Knowledge Representation                                                                               Human demonstration...
Indexing of the Atlas information from 3D perception                                              Real-world scenario     ...
Algorithm Description                                                            Algorithm Description Machine Vision and ...
3D Object Recognition                                                        3D Object RecognitionMachine Vision and Perce...
Online Recognition Phase                                                 Online Recognition PhaseMachine Vision and Percep...
Online Recognition Phase                                                 Online Recognition PhaseMachine Vision and Percep...
MVP   Machine Vision and Perception Group @ TUM                                                  Example of the System in ...
How to reconstruct 3D under poor texture                                            condition?Machine Vision and Perceptio...
What can we do if the texture information is                                            almost non-existent?Machine Vision...
Intensity-Based MethodsMachine Vision and Perception Group @ TUM                                                         ...
Intensity-Based Bundle Adjustment                                                                                        ...
Reconstruction ExampleMachine Vision and Perception Group @ TUM                                            Works well und...
Estimation of Physical Properties                                            in Predict-Act-Perceive loopMachine Vision an...
Machine Vision and Perception Group @ TUM                                            Knowledge Representation in WP5      ...
Defining Foreground: Shape Matching                      Indexing to the Atlas database needs to be                      e...
Object ContainerMachine Vision and Perception Group @ TUM                                            Orientation          ...
Functionality Map                                            Location AreasMachine Vision and Perception Group @ TUM      ...
Basic Experiments:                                            Location Areas (Tracking Data)Machine Vision and Perception ...
Basic Experiments:                                            Functionality Maps (Tracking Data)Machine Vision and Percept...
MVP   Machine Vision and Perception Group @ TUM                                            Trajectories                   ...
Upcoming SlideShare
Loading in …5
×

Mvp

3,131 views

Published on

Research projects at my TUM group - MVP

Published in: Technology, Business
  • Be the first to comment

Mvp

  1. 1. Research of the MVP Group http://mvp.visual-navigation.com Visual navigation Perception for manipulation Biologically motivatedMachine Vision and Perception Group @ TUM perception Rigid and Deformable The Machine Registration Vision and Perception Group @TUM works on the aspects of visual perception and control in medical, mobile, and HCI applications Photogrammetric Visual Action Analysis Exploration of physical monocular reconstruction object propertiesMVP
  2. 2. Visual Localization for hybrid Environmental ModelingMachine Vision and Perception Group @ TUM Real-Time Localization The high accuracy allows direct stiching of images along the trajectory without bundle adjustment Resulting hybrid (appearance and geometry) model allows path planning and prediction of sensor views for e.g. attentionMVP research
  3. 3. MVP Machine Vision and Perception Group @ TUM Example of a hybrid model reconstruction
  4. 4. Result of a visual localization 6DoF pose of the robot is calculated fom single camera viewMachine Vision and Perception Group @ TUMMVP
  5. 5. Reconstruction of a hybrid (appearance/geometry) modelMachine Vision and Perception Group @ TUMMVP
  6. 6. Visual HomingMachine Vision and Perception Group @ TUM The system is capable of registration to previously observed trajectoriesMVP
  7. 7. Biologically Motivated Navigation Obstacle avoidanceMachine Vision and Perception Group @ TUM Decoupling of Rotation and Translation Perception-based loop closureMVP
  8. 8. MVP Machine Vision and Perception Group @ TUM Work on Optimal Sensor Models
  9. 9. Knowledge Representation Human demonstrationMachine Vision and Perception Group @ TUM • Atlas: Atlas parametric grasp actions and – Long-term memory shape description points handling – Experience of the system • Working memory: – Short-term memory Mapping of – Experience grounded in a Knowledge given environment Foreground Experience segmentation abstraction • Temporal handling information background foreground Episodic buffer model model of actionsMVP Working Memory
  10. 10. Indexing of the Atlas information from 3D perception Real-world scenario Real-world scenarioMachine Vision and Perception Group @ TUM scene setupMVP input point cloud recognized models
  11. 11. Algorithm Description Algorithm Description Machine Vision and Perception Group @ TUM (Model Preprocessing Phase) (Model Preprocessing Phase) n2 • For all pairs of surflets at  distance d insert the triple  p2   ,  , ∢ n1 , n 2  n1 d p1 plus a pointer to its model in a hash-table. • Do this for all models using theMVP same hash-table.
  12. 12. 3D Object Recognition 3D Object RecognitionMachine Vision and Perception Group @ TUM • Challenges ― Incomplete Data ― Noise and outliersMVP
  13. 13. Online Recognition Phase Online Recognition PhaseMachine Vision and Perception Group @ TUM • For each model surflet pair  p1 , n1  , p2 , n2      in the hash-table cell: n1 p1 Compute the rigid transform T that p2  p ,alignsp , n  best n  ,    n2 1 1 2 2 to  p 1 , n1  , p2 , n2 MVP model hash-table
  14. 14. Online Recognition Phase Online Recognition PhaseMachine Vision and Perception Group @ TUM • Check if T(M) matches the scene. M is the model of n1  p1 , n1  , p2 , n2      p1 n1 p1   p2 p2 n2 n2 MVP • If yes, accept T(M). model hash-table
  15. 15. MVP Machine Vision and Perception Group @ TUM Example of the System in Action
  16. 16. How to reconstruct 3D under poor texture condition?Machine Vision and Perception Group @ TUM Problem: texture information is more sparseMVP 16
  17. 17. What can we do if the texture information is almost non-existent?Machine Vision and Perception Group @ TUM → photogrammetric approachMVP 17
  18. 18. Intensity-Based MethodsMachine Vision and Perception Group @ TUM  Established 3D reconstruction methods:  Rely on presence of many reliable features  May perform very poor if this prerequisite is not met  Medical images are typically problematic  Solution approach: Intensity based methods  Do not exclusively rely on salient image regions  Shaded regions, e.g., also contain information  Use all information that is thereMVP
  19. 19. Intensity-Based Bundle Adjustment  Regular Bundle Adjustment equation:Machine Vision and Perception Group @ TUM min a , b ∑i=1 ∑m=1 d (Q (a j , b i ) , x i , j ) n j  Finds camera parameters a and point coordinates b  Minimizes feature reprojection error min a , b ∑n ∑m d ( I j (Q ( a j , b i )) , I 0 ( p i )) i=1 j=1  Intensity-Based Bundle Adjustment:  Uses a regularizer: b now describes a smooth surface  Apart from that: Same, but minimizes intensity errorMVP
  20. 20. Reconstruction ExampleMachine Vision and Perception Group @ TUM Works well under static lighting conditions and roughly Lambertian surfacesMVP Ruepp and Burschka. Fast recovery of weakly textured surfaces from monocular image sequences. ACCV 2010
  21. 21. Estimation of Physical Properties in Predict-Act-Perceive loopMachine Vision and Perception Group @ TUM Estimation of the Center of mass Estimation of Mass and Friction Force Estimation of Mass Distribution predict act perceive F r τMVP simulator
  22. 22. Machine Vision and Perception Group @ TUM Knowledge Representation in WP5 Mapping of KnowledgeMVP
  23. 23. Defining Foreground: Shape Matching Indexing to the Atlas database needs to be extended to object classes -> deformable shape registration needed (ACCV2010 - oral) 23
  24. 24. Object ContainerMachine Vision and Perception Group @ TUM Orientation Max. allowed acceleration Mass Center of gravityMVP
  25. 25. Functionality Map Location AreasMachine Vision and Perception Group @ TUM Connection Properties: Pushed vs. lifted object Arbitrary movement vs. movement with a goal Connection relevance Velocity constraint during pick-upMVP Grasp taxonomy Approach vector
  26. 26. Basic Experiments: Location Areas (Tracking Data)Machine Vision and Perception Group @ TUMMVP
  27. 27. Basic Experiments: Functionality Maps (Tracking Data)Machine Vision and Perception Group @ TUM Milk carton Cup - HandleMVP Red = goal, green = push, magenta = arbitrary
  28. 28. MVP Machine Vision and Perception Group @ TUM Trajectories Entire System: Exemplary

×