Workshop ICDL







Since birth humans show a strong predisposition to social interaction, which is supported by their developing visual perception skills. An important role is played by the vision of others’ actions and movements. For instance, a natural predisposition to detect biological motion is present from birth enabling infants to detect interacting agents [1]. More refined abilities — as the understanding and anticipation of others’ actions [2], [3] — progressively develop with age, leading, in a few years, to a full capability of interaction based on mutual understanding, joint coordination and collaboration.


Unfolding the theory of visual motion perception development is one of the main challenges of neuro and cognitive science researches. Over the last decades, the artificial vision community has shown an increasing interest for these theories, stimulated by the ambition of providing artificial agents with comparable perception capabilities (e.g. [4], [5]). The topics of the workshop will refer to the possible interconnections between human and machine vision in a developmental perspective.


The purpose of the meeting will be two-fold: on the one hand the discussion will focus on how the development of  human visual perception might inspire the development of novel methods for artificial vision and social robotics; on the other hand, it will evaluate how the implementation of machine vision methods could help understanding human social development. The ambition of this workshop is to stimulate the discussion on these challenging topics, bridging Computer Vision, Developmental Science and Robotics, with an eye to potential applications in the field of human-robot interactions and rehabilitation.




[1] F. Simion, L. Regolin, and H. Bulf, “A predisposition for biological motion in the newborn baby.,” Proc. Natl. Acad. Sci. U. S. A., vol. 105, no. 2, pp. 809–13, Jan. 2008.
[2] T. Falck-Ytter, G. Gredebäck, and C. von Hofsten, “Infants predict other people’s action goals.,” Nat. Neurosci., vol. 9, no. 7, pp. 878–879, Jul. 2006.
[3] Y. Kanakogi and S. Itakura, “Developmental correspondence between action prediction and motor ability in early infancy.,” Nat. Commun., vol. 2, no. May, p. 341, Jan. 2011.
[4] N. Noceti, A. Sciutti, F. Rea, F. Odone, A. Verri, and G. Sandini, “Biological motion understanding for human-robot interaction,” in Vision for Language and Manipulation BMVA symposium, London, UK, July, 11th, 2014, 2014.
[5] N. Noceti, A. Sciutti, F. Rea, F. Odone, and G. Sandini, “Estimating human actions affinities across views,” in International Conference on Computer Vision Theory and Applications (VISAPP), 2015.