Invariant visual object recognition: a model, with lighting invariance.

How are invariant representations of objects formed in the visual cortex? We describe a neurophysiological and computational approach which focusses on a feature hierarchy model in which invariant representations can be built by self-organizing learning based on the statistics of the visual input. The model can use temporal continuity in an associative synaptic learning rule with a short term memory trace, and/or it can use spatial continuity in Continuous Transformation learning. The model of visual processing in the ventral cortical stream can build representations of objects that are invariant with respect to translation, view, size, and in this paper we show also lighting. The model has been extended to provide an account of invariant representations in the dorsal visual system of the global motion produced by objects such as looming, rotation, and object-based movement. The model has been extended to incorporate top-down feedback connections to model the control of attention by biased competition in for example spatial and object search tasks. The model has also been extended to account for how the visual system can select single objects in complex visual scenes, and how multiple objects can be represented in a scene.     

Learning invariant object recognition in the visual system with continuous transformations

The cerebral cortex utilizes spatiotemporal continuity in the world to help build invariant representations. In vision, these might be representations of objects. The temporal continuity typical of objects has been used in an associative learning rule with a short-term memory trace to help build invariant object representations. In this paper, we show that spatial continuity can also provide a basis for helping a system to self-organize invariant representations. We introduce a new learning paradigm “continuous transformation learning” which operates by mapping spatially similar input patterns to the same postsynaptic neurons in a competitive learning system. As the inputs move through the space of possible continuous transforms (e.g. translation, rotation, etc.), the active synapses are modified onto the set of postsynaptic neurons. Because other transforms of the same stimulus overlap with previously learned exemplars, a common set of postsynaptic neurons is activated by the new transforms, and learning of the new active inputs onto the same postsynaptic neurons is facilitated. We demonstrate that a hierarchical model of cortical processing in the ventral visual system can be trained with continuous transform learning, and highlight differences in the learning of invariant representations to those achieved by trace learning.     

Early development of visual recognition

The most important ability of the human vision is object recognition, yet it is exactly the less understood aspect of the vision system. Computational models have been helpful in progressing towards an explanation of this obscure cognitive ability, and today it is possible to conceive more refined models, thanks to the new availability of neuroscientific data about the human visual cortex. This work proposes a model of the development of the object recognition capability, under a different perspective with respect to the most common approaches, with a precise theoretical epistemology. It is assumed that the main processing functions involved in recognition are not genetically determined and hardwired in the neural circuits, but are the result of interactions between epigenetic influences and the basic neural plasticity mechanisms. The model is organized in modules related with the main visual biological areas, and is implemented mainly using the LISSOM architecture, a recent self-organizing algorithm closely reflecting the essential behavior of cortical circuits.     

A feature extractor for curvilinear patterns: a design suggested by the mammalian visual system

Summary A feature extractor for a pattern recognizer which can effectively process curvilinear drawings has been synthesized and simulated on a digital computer.The design of the network was suggested by the visual system of higher animals — especially the structure of the receptive fields of cortical neurons. This feature extractor is a multilayered parallel network composed of analog threshold elements. It consists of six layers in cascade. The first layer is a two-dimensional array of photoreceptors. The second layer is a contrast-detecting layer, each element of which has an on-center-type receptive field. The third one is a line-detecting layer. An element of this layer corresponds to a simple cortical cell, and responds to lines whose orientation is proper for the element. Each element has a receptive field consisting of an elongated excitatory region flanked on either side by inhibitory regions. The fourth layer is also a line-detecting layer, but each element, which corresponds to a complex cell, is not sensitive to the exact position of the line. An element of the fifth layer, which may correspond to a hypercomplex cell, responds when the line detected in the preceeding layer is curved. In the final layer, the curvature of the line is detected regardless of the orientation of the line, that is, an element of this layer gives an output approximately proportional to the curvature of the line presented in its receptive field.     

Real-time unconstrained object recognition: a processing pipeline based on the mammalian visual system

The mammalian visual system is still the gold standard for recognition accuracy, flexibility, efficiency, and speed. Ongoing advances in our understanding of function and mechanisms in the visual system can now be leveraged to pursue the design of computer vision architectures that will revolutionize the state of the art in computer vision.     

Mechanisms of visual object recognition studied in monkeys

Cells in area TE of the inferotemporal cortex of the monkey brain selectively respond to various moderately complex object-features, and those responding to similar features cluster in a columnar region elongated vertical to the cortical surface. Although cells within a column respond to similar features, their selectivity is not identical. The data of optical imaging in TE have suggested that the borders between neighboring columns are not discrete but columns representing related features overlap one another. We have also found, by training adult monkeys for discrimination of a specific set of shapes, that such a long-term training increases the proportion of TE cells responding to the shapes used in the training even in the adult. The data suggested that TE plays important roles in discrimination of complex shapes and in visual expert learning of discriminating a certain class of objects in the adult.     

A theory for the development of feature detecting cells in visual cortex

Passive modification of the strength of synaptic junctions that results in the construction of internal mappings with some of the properties of memory is shown to lead to the development of Hubel-Wiesel type feature detectors in visual cortex. With such synaptic modification a cortical cell can become committed to an arbitrary but repeated external pattern, and thus fire every time the pattern is presented even if that cell has no genetic pre-disposition to respond to the particular pattern. The additional assumption of lateral inhibition between cortical cells severely limits the number of cells which respond to one pattern as well as the number of patterns that are picked up by a cell. The introduction of a simple neural mapping from the visual field to the lateral geniculate leads to an interaction between patterns which, combined with our assumptions above, seems to lead to a progression of patterns from column to column of the type observed by Hubel and Wiesel in monkey.     

The effect of the visual context in the recognition of symbolic gestures

Background

To investigate, by means of fMRI, the influence of the visual environment in the process of symbolic gesture recognition. Emblems are semiotic gestures that use movements or hand postures to symbolically encode and communicate meaning, independently of language. They often require contextual information to be correctly understood. Until now, observation of symbolic gestures was studied against a blank background where the meaning and intentionality of the gesture was not fulfilled.

Methodology/Principal Findings

Normal subjects were scanned while observing short videos of an individual performing symbolic gesture with or without the corresponding visual context and the context scenes without gestures. The comparison between gestures regardless of the context demonstrated increased activity in the inferior frontal gyrus, the superior parietal cortex and the temporoparietal junction in the right hemisphere and the precuneus and posterior cingulate bilaterally, while the comparison between context and gestures alone did not recruit any of these regions.

Conclusions/Significance

These areas seem to be crucial for the inference of intentions in symbolic gestures observed in their natural context and represent an interrelated network formed by components of the putative human neuron mirror system as well as the mentalizing system.     

One-trial visual recognition in cats: the role of the rhinal cortex

Visual recognition memory in primates is mediated at least in part by the perirhinal and entorhinal (i.e., rhinal) cortices. To examine the role of these structures in cats' visual recognition memory, we performed combined electrolytic rhinal (perirhinal and entorhinal) lesions in a group of cats trained in visual delayed matching-to-sample with trial-unique objects in the modified Wisconsin General Testing Apparatus. Sham-operated and intact cats were used as control groups. Cats with rhinal lesions did not differ from the control sham-operated and unoperated groups in initial learning of the rules of the task; difference between experimental and control groups under conditions of minimum 5-sec delay was nonsignificant as well. However, significant difference between experimental and control groups was revealed under conditions of testing with 10-sec delay. This finding suggests a disorder in the visual recognition memory.