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.

Autonomous Learning of Robust Visual Object Detection & Identification on a Humanoid #icdl/epirob2012


Published on

n this work we introduce a technique for a hu- manoid robot to autonomously learn the representations of objects in its visual environment. Our approach involves feature- based segmentation of the images followed by learning to identify the object using Cartesian Genetic Programming. The learned identification is able to provide robust and fast segmentation of the objects, without using features. To allow for autonomous learning an attention mechanism is coupled with the training process. We showcase our system on a humanoid robot.

Published in: Technology
  • Be the first to comment

  • Be the first to like this

Autonomous Learning of Robust Visual Object Detection & Identification on a Humanoid #icdl/epirob2012

  1. 1. Jürgen ’Juxi’ Leitner S. Harding, P. Chandrashekhariah M. Frank, G. Spina, A. Förster, J. Triesch, J. Schmidhuber idsia / usi / supsi, machine intelligence, fias autonomous learning of robust visual objectdetection & identification on a humanoid #icdl/epirob 2012
  2. 2. our iCub setup is formanipulation
  3. 3. thanks to G. Metta and IIT for this picture visual perception
  4. 4. thechallenge
  5. 5. IDSIA’s three partsHarding et al., GPTP 2012, Leitner et al., IROS 2012 Leitner et al., BICA 2012 Leitner et al., ICDL 2012
  6. 6. currentcv approaches
  7. 7. detecting objects Harding et al., GPTP 2012
  8. 8. our learning approach
  9. 9. INP INP INP + dilate min avg cartesian genetic programming
  10. 10. INP INP INP + dilate min avgFunc,on" 3"Connec,on"1" "#2"Connec,on"2"A"real"number" "#1" 4.3" cartesian genetic """ " programming
  11. 11. icImage GreenTeaBoxDetector::runFilter() {! icImage node0 = InputImages[6];! icImage node1 = InputImages[1];! icImage node2 = node0.absdiff(node1);! icImage node5 = node2.SmoothBilateral(11);! icImage node12 = InputImages[0];! icImage node16 = node12.Sqrt();! icImage node33 = node16.erode(6);! icImage node34 = node33.log();! icImage node36 = node34.min(node5);! icImage node49 = node36.Normalize(); //cleanup ... icImage out = node49.threshold(230.7218f);! return out; }detection
  12. 12. detect
  13. 13. detect
  14. 14. BUT: supervised learning approach
  15. 15. FIAScollaborationsaliencymap feature segmentation
  16. 16. presegmentation
  17. 17. combinedapproach
  18. 18. MoBeEframework Frank et al., ICINCO, 2012.
  19. 19. conclusion salient object detection + rough feature-based segmentation = automatic training set generation for cgp-based, robust filter-learning
  20. 20. thanks for video at