Improving Robot Vision Models for Object Detection Through Interaction #ijcnn2014

Jürgen `Juxi’ Leitner

Improving Robot Vision Through Interaction

Dalle Molle Institute for AI (IDSIA)

#ijcnn 2014

manipulation

our iCub setup is for

object manipulationtowards learning

http://robotics.idsia.ch/

generationmotionStollenga et al, 2013

Shak

ey 2

013

Win

ner

MoBeEframework Frank et al., 2012.

MoBeE

!

transferringspatial perception

Leitner IROS 2012/IJCNN 2013

!

perceptionvisual

thanks to G. Metta and IIT for this picture

!

objectsdetecting

Harding et al., 2013Leitner et al., ICDL 2012, ARS 2012, BICA 2012, CEC 2013

cartesian genetic programming

+ min dilate avg INP INP INP

[Leitner et al, 2012a/b, Harding et al., 2013]

detection

! icImage* BlueCupFilter::runFilter() { icImage* node43 = InputImages[4]; icImage* node49 = node43->LocalAvg(15); ! icImage* out = node49->threshold(81.532f); return out; }

manipulation for improved perception

actions

extracting information

improveddetection

detection

! icImage* BlueCupFilter::runFilter() { icImage* node0 = InputImages[4].Exp(); icImage* node5 = InputImages[0]; icImage* node16 = node0->Gabor(-8,14,1,13); icImage* node17 = InputImages[4]->LocalAvg(6); icImage* node18 = node16->Laplace(5); icImage* node19 = node5->Sobel(13,9); icImage* node24 = node17->Erode(5); icImage* node28 = node19->Min(node18); icImage* node29 = node28->Min(node24); icImage* node41 = node29->LocalAvg(7); icImage* node49 = node41->LocalMax(7); ! icImage* out = node49->threshold(68.032f); return out; }

resulting

!

workfuture

eye-hand coordination

different object representations

online, continuous learning

for listeningthanks

[email protected] http://Juxi.net/projects