Discovering containment: from infants to machines

Current artificial learning systems can recognize thousands of visual categories, or play Go at a champion"s level, but cannot explain infants learning, in particular the ability to learn complex concepts without guidance, in a specific order. A notable example is the category of 'containers' and the notion of containment, one of the earliest spatial relations to be learned, starting already at 2.5 months, and preceding other common relations (e.g., support). Such spontaneous unsupervised learning stands in contrast with current highly successful computational models, which learn in a supervised manner, that is, by using large data sets of labeled examples. How can meaningful concepts be learned without guidance, and what determines the trajectory of infant learning, making some notions appear consistently earlier than others?

[1]  Aimee E. Stahl,et al.  Observing the unexpected enhances infants’ learning and exploration , 2015, Science.

[2]  Michael J. Black,et al.  A Fully-Connected Layered Model of Foreground and Background Flow , 2013, 2013 IEEE Conference on Computer Vision and Pattern Recognition.

[3]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[4]  Albert Yonas,et al.  Infants' sensitivity to pictorial depth cues: a review and meta-analysis of looking studies. , 2012, Infant behavior & development.

[5]  Shimon Ullman,et al.  The chains model for detecting parts by their context , 2010, 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[6]  Marianella Casasola,et al.  Learning to form a spatial category of tight-fit relations: how experience with a label can give a boost. , 2009, Developmental psychology.

[7]  H. Neumann,et al.  Object Segmentation from Motion Discontinuities and Temporal Occlusions–A Biologically Inspired Model , 2008, PloS one.

[8]  Marianella Casasola,et al.  The Development of Infants' Spatial Categories , 2008 .

[9]  David Silverstein,et al.  The Case of the Containers , 2007 .

[10]  Daniel P. Huttenlocher,et al.  Spatial priors for part-based recognition using statistical models , 2005, 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05).

[11]  R. Baillargeon,et al.  When the ordinary seems unexpected: evidence for incremental physical knowledge in young infants , 2005, Cognition.

[12]  R. Baillargeon,et al.  Detecting continuity violations in infancy: a new account and new evidence from covering and tube events , 2005, Cognition.

[13]  David G. Lowe,et al.  Distinctive Image Features from Scale-Invariant Keypoints , 2004, International Journal of Computer Vision.

[14]  Susan J. Hespos,et al.  Conceptual precursors to language , 2004, Nature.

[15]  M. Casasola,et al.  Six-month-old infants' categorization of containment spatial relations. , 2003, Child development.

[16]  M. Casasola,et al.  Infant categorization of containment, support and tight‐fit spatial relationships , 2002 .

[17]  Susan J. Hespos,et al.  Reasoning about containment events in very young infants , 2001, Cognition.

[18]  R. Baillargeon,et al.  2.5-Month-Old Infants' Reasoning about When Objects Should and Should Not Be Occluded , 1999, Cognitive Psychology.

[19]  Jitendra Malik,et al.  Normalized cuts and image segmentation , 1997, Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[20]  Laurent Najman,et al.  Geodesic Saliency of Watershed Contours and Hierarchical Segmentation , 1996, IEEE Trans. Pattern Anal. Mach. Intell..

[21]  Scott P. Johnson,et al.  Perception of object unity in young infants: The roles of motion, depth, and orientation , 1996 .

[22]  Michal Irani,et al.  Computing occluding and transparent motions , 1994, International Journal of Computer Vision.

[23]  P. Burt,et al.  Mechanisms for isolating component patterns in the sequential analysis of multiple motion , 1991, Proceedings of the IEEE Workshop on Visual Motion.

[24]  Shmuel Peleg,et al.  Motion based segmentation , 1990, [1990] Proceedings. 10th International Conference on Pattern Recognition.

[25]  E. Spelke,et al.  Object Perception in Infancy: Interaction of Spatial and Kinetic Information for Object Boundaries. , 1989 .

[26]  Berthold K. P. Horn,et al.  Direct methods for recovering motion , 1988, International Journal of Computer Vision.

[27]  R. Kaufmann-Hayoz,et al.  Kinetic contours in infants' visual perception. , 1986, Child development.

[28]  A. Yonas,et al.  Infants' sensitivity to accretion and deletion of texture as information for depth at an edge. , 1984, Child development.

[29]  E. Spelke,et al.  Perception of partly occluded objects in infancy , 1983, Cognitive Psychology.

[30]  F. J. Langdon,et al.  The Child's Conception of Space , 1967 .

[31]  Guigang Zhang,et al.  Deep Learning , 2016, Int. J. Semantic Comput..

[32]  Shimon Ullman,et al.  Learning to perceive coherent objects , 2013, CogSci.

[33]  Ernest Davis,et al.  Reasoning from Radically Incomplete Information: The Case of Containers , 2013 .

[34]  Mitchell Rabinowitz,et al.  Representation, memory, and development : essays in honor of Jean Mandler , 2012 .

[35]  B. Schiele,et al.  Combined Object Categorization and Segmentation With an Implicit Shape Model , 2004 .

[36]  Scott P. Johnson,et al.  Perception of kinetic illusory contours by two-month-old infants. , 2002, Child development.

[37]  Susan J. Hespos,et al.  Conceptual development in infancy: The case of containment , 2002 .

[38]  A. Needham Object recognition and object segregation in 4.5-month-old infants. , 2001, Journal of experimental child psychology.

[39]  A. Needham,et al.  Infants' use of prior experiences with objects in object segregation: implications for object recognition in infancy. , 1999, Advances in child development and behavior.

[40]  R. Baillargeon,et al.  Effects of prior experience on 4.5-month-old infants' object segregation , 1998 .

[41]  Oliver Braddick,et al.  Binocularity in infancy , 1996, Eye.

[42]  Michael J. Black,et al.  The Robust Estimation of Multiple Motions: Parametric and Piecewise-Smooth Flow Fields , 1996, Comput. Vis. Image Underst..

[43]  Ad W. Smitsman,et al.  Infants' Perception of Dynamic Relations Between Objects: Passing Through or Support? , 1995 .

[44]  R. Baillargeon,et al.  The Development of Young Infants' Intuitions about Support , 1992 .

[45]  Elizabeth S. Spelke,et al.  Principles of Object Perception , 1990, Cogn. Sci..

[46]  Alessandro Verri,et al.  Motion Segmentation from Optical Flow , 1989, Alvey Vision Conference.

[47]  A Yonas,et al.  Infants’ sensitivity to kinetic information for three-dimensional object shape , 1988, Perception & psychophysics.

[48]  W. B. Thompson,et al.  Relative motion: Kinetic information for the order of depth at an edge , 1987, Perception & psychophysics.

[49]  Malcolm Acock,et al.  Vision: A Computational Investigation into the Human Representation and Processing of Visual Information. By David Marr , 1985 .