We present a method for biologically-inspired object recognition with one-shot learning of object appearance. We use a computationally efficient model of V1 keypoints to select object parts with the highest information content and model their surroundings using simple colour features. This map-like representation is fed into a dynamical neural network which performs pose, scale and translation estimation of the object given a set of previously observed object views. We demonstrate the feasibility of our algorithm for cognitive robotic scenarios and evaluate classification performance on a dataset of household items. © 2014 Springer International Publishing Switzerland.
CITATION STYLE
Lomp, O., Terzić, K., Faubel, C., Du Buf, J. M. H., & Schöner, G. (2014). Instance-based object recognition with simultaneous pose estimation using keypoint maps and neural dynamics. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 8681 LNCS, pp. 451–458). Springer Verlag. https://doi.org/10.1007/978-3-319-11179-7_57
Mendeley helps you to discover research relevant for your work.