Biologically Inspired Model for Visual Cognition Achieving Unsupervised Episodic and Semantic Feature Learning

Abstract

Recently, many biologically inspired visual computational models have been proposed. The design of these models follows the related biological mechanisms and structures, and these models provide new solutions for visual recognition tasks. In this paper, based on the recent biological evidence, we propose a framework to mimic the active and dynamic learning and recognition process of the primate visual cortex. From principle point of view, the main contributions are that the framework can achieve unsupervised learning of episodic features (including key components and their spatial relations) and semantic features (semantic descriptions of the key components), which support higher level cognition of an object. From performance point of view, the advantages of the framework are as follows: 1) learning episodic features without supervision-for a class of objects without a prior knowledge, the key components, their spatial relations and cover regions can be learned automatically through a deep neural network (DNN); 2) learning semantic features based on episodic features-within the cover regions of the key components, the semantic geometrical values of these components can be computed based on contour detection; 3) forming the general knowledge of a class of objects-the general knowledge of a class of objects can be formed, mainly including the key components, their spatial relations and average semantic values, which is a concise description of the class; and 4) achieving higher level cognition and dynamic updating-for a test image, the model can achieve classification and subclass semantic descriptions. And the test samples with high confidence are selected to dynamically update the whole model. Experiments are conducted on face images, and a good performance is achieved in each layer of the DNN and the semantic description learning process. Furthermore, the model can be generalized to recognition tasks of other objects with learning ability.

MHTN: Modal-Adversarial Hybrid Transfer Network for Cross-Modal Retrieval

“…Recently, a few works have been proposed for incorporating various information into common representation learning, such as semi-supervised and sparse regularizations [5], local group based priori [22], and semantic hierarchy [23]. Inspired by the considerable improvement by DNN in many single-modal tasks such as image classification [15] and object recognition [24], researchers have made great efforts to apply DNN to cross-modal retrieval as [6], [9], [25], [26]. For example, Ngiam et al [9] propose bimodal deep autoencoder, which is an extension of restricted Boltzmann machine (RBM).…”

Object Detection Recognition and Robot Grasping Based on Machine Learning: A Survey

“…The low sampling efficiency of reinforcement learning makes training difficult, and a reasonable reward function and network structure need to be designed to achieve better results. Robot [163][164][165][166] Computer vision [167][168][169][170][171] Data analysis [23,172,173] Self-supervised…”

A Survey of Robotics Control Based on Learning-Inspired Spiking Neural Networks

“…Therefore, similarly, networks of this type are usually adopted for low-level sensory acquisition in robotic systems, such as vision (Perrinet et al, 2004 ), tactile sensing (Rochel et al, 2002 ), and olfaction (Cassidy and Ekanayake, 2006 ). For example, inspired by the structures and principles of primate visual cortex, Qiao et al ( 2014 , 2015 , 2016 ) enhanced the feed-forward models including Hierarchical Max Pooling (HAMX) model and Convolutional Deep Belief Network (CDBN) with memory, association, active adjustment, semantic and episodic feature learning ability etc., and achieved good results in visual recognition task.…”