Abstract
Handling objects or interacting with a human user about objects on a shared tabletop requires that objects be identified after learning from a small number of views and that object pose be estimated. We present a neurally inspired architecture that learns object instances by storing features extracted from a single view of each object. Input features are color and edge histograms from a localized area that is updated during processing. The system finds the best-matching view for the object in a novel input image while concurrently estimating the object's pose, aligning the learned view with current input. The system is based on neural dynamics, computationally operating in real time, and can handle dynamic scenes directly off live video input. In a scenario with 30 everyday objects, the system achieves recognition rates of 87.2% from a single training view for each object, while also estimating pose quite precisely. We further demonstrate that the system can track moving objects, and that it can segment the visual array, selecting and recognizing one object while suppressing input from another known object in the immediate vicinity. Evaluation on the COIL-100 dataset, in which objects are depicted from different viewing angles,...Continue Reading
References
Aug 3, 1977·Biological cybernetics·S Amari
Feb 1, 1992·Journal of Neurophysiology·C Prablanc, O Martin
Jan 1, 1980·Biological cybernetics·K Fukushima
Jun 6, 1996·Nature·S ThorpeC Marlot
Feb 17, 2000·Neuron·M Riesenhuber, T Poggio
Apr 9, 2002·Neural Computation·Laurenz Wiskott, Terrence J Sejnowski
Nov 24, 2004·Neural Networks : the Official Journal of the International Neural Network Society·Junmei Zhu, Christoph von der Malsburg
May 25, 2005·Journal of Experimental Psychology. Learning, Memory, and Cognition·Andrew Hollingworth
Jun 13, 2006·Neural Computation·Geoffrey E HintonYee-Whye Teh
Jan 17, 2007·IEEE Transactions on Pattern Analysis and Machine Intelligence·Thomas SerreTomaso Poggio
May 27, 2008·Neural Networks : the Official Journal of the International Neural Network Society·Christian Faubel, Gregor Schöner
Jan 17, 2009·Journal of Vision·Philipp WolfrumChristoph von der Malsburg
Aug 4, 2011·Frontiers in Psychology·Michael L Mack, Thomas J Palmeri
Citations
Feb 13, 2020·Attention, Perception & Psychophysics·Raul GriebenGregor Schöner
Dec 6, 2019·Frontiers in Neurorobotics·Jan TekülveGregor Schöner
Oct 15, 2021·Cognitive Science·Mathis RichterGregor Schöner