1 / 25

Better Vision through

Better Vision through. Poking. Poking. Poking. Paul Fitzpatrick Humanoid Robotics Group MIT AI Lab. Strategies for Sensing. Sensing capability is finite Many choices in how to allocate it Smart choices simplify processing. Example 1: Human Foveation. (Adapted from V. Tucker ‘00).

nerina
Télécharger la présentation

Better Vision through

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Better Vision through Poking Poking Poking Paul Fitzpatrick Humanoid Robotics Group MIT AI Lab

  2. Strategies for Sensing • Sensing capability is finite • Many choices in how to allocate it • Smart choices simplify processing

  3. Example 1: Human Foveation (Adapted from V. Tucker ‘00)

  4. Example 1: Falcon Foveation (Adapted from V. Tucker ‘00)

  5. Example 1: Robot Foveation

  6. Example 2: Shape from Probing (Figure from E. Paulos ‘99)

  7. Example 2: Shape from Probing (Figure from E. Paulos ‘99)

  8. Ow! Image Segmentation • Image segmentation is subtle, ambiguous Physical poking is direct, to the point

  9. Active Segmentation • Unsure of an object’s boundaries? • Poke it gently • Thump it savagely • Try to put your arm/hand/flipper beside it • Try to put your arm/hand/flipper behind it • Move your head for a different view • Get help

  10. A Simple Scene?

  11. A Simple Scene? Edges of table and cube overlap Color of cube and table are poorly separated Cube has misleading surface pattern

  12. Active Segmentation

  13. Active Segmentation

  14. Result No confusion between cube and own texture No confusion between cube and table

  15. Anatomy of a Poke Begin Find hand Sweep Contact! Withdraw

  16. Head (7 DOFs) Right arm (6 DOFs) Left arm (6 DOFs) Torso (3 DOFs) Stand (0 DOFs) The Robot

  17. The Arm shoulder (b) shoulder (a) elbow (b) elbow (a) wrist (a) wrist (b)

  18. right eye pan eye tilt left eye pan top differential bottom differential neck pan neck tilt The Head

  19. Tracing Cause and Effect • Goal: to relate robot and human action without prior knowledge of visual appearance • Determine appearance of own arm in motion • Follow the causal chain outwards to determine the appearance of actions on objects • Then follow the chain “up” a human’s arm when they move an object after the robot

  20. Tracing Cause and Effect

  21. Locating Arm without Appearance Model • Shake it • Correlate commanded motion with optic flow • Ignore uncorrelated motion (Giorgio Metta)

  22. Locating Arm without Appearance Model Optical flow Maximum Segmented regions (Giorgio Metta)

  23. Training Visual Predictor (Giorgio Metta)

  24. Then… • Start poking things!

  25. Things To Do • Segment completely visually ambiguous scenes • Characterize non-rigid objects – should lead to pragmatic, realistic object model

More Related