V4 Neural Network Model for Shape-Based Feature Extraction and Object Discrimination
Cognitive ComputationPublished 13 October 2015
Hui Wei, Dong Zheng
Citations12
SJR quartileQ1
SJR score0.84
SNIP1.29
Generate an AI Snapshot to get a quick, structured summary of this paper.
Study Snapshot
ObjectiveStudy objective
MethodsResearch methodology
PopulationPopulation studied
Sample sizeSample sizes
OutcomesStudy outcomes here
ResultsStudy results comes here
LimitationsResearch study limitations comes here
A concise AI-generated summary of the paper will appear here once you click Generate AI Snapshot.
TL;DR
A novel neural network model of V4 for shape-based feature extraction and other vision tasks and shows a successful attempt to construct a computation model of visual object recognition in the brain.
Abstract
Visual area V4 plays an important role in the neural mechanism of shape recognition. V4 neurons exhibit selectivity for the orientation and curvature of boundary fragments. In this paper, we propose a novel neural network model of V4 for shape-based feature extraction and other vision tasks. The low-level layers of the model consist of computational units simulating simple cells and complex cells in the primary visual cortex. These layers extract preliminary visual features including edges and orientations. The V4 computational units calculate the entropy of the extracted features as a measure of visual saliency. The features around salient points are then selected and encoded with a layer of restricted Boltzmann machine to generate an intermediate representation of object shapes. The model is evaluated in shape distinction, feature detection, feature matching, and object discrimination experiments. The results demonstrate that this model generates discriminative local representation of object shapes. It shows a successful attempt to construct a computation model of visual object recognition in the brain.
Keywords
Computer ScienceNeuroscience
Communications of the ACMImageNet classification with deep convolutional neural networks
75,709 Citations2017Alex Krizhevsky, Ilya Sutskever +1 more
A large, deep convolutional neural network was trained to classify the 1.2 million high-resolution images in the ImageNet LSVRC-2010 contest into the 1000 different classes and employed a recently developed regularization method called "dropout" that proved to be very effective.
Object recognition from local scale-invariant features
16,195 Citations1999David Lowe
Experimental results show that robust object recognition can be achieved in cluttered partially occluded images with a computation time of under 2 seconds.
Annual Review of NeuroscienceNeural Mechanisms of Selective Visual Attention
8,229 Citations1995Robert Desimone, John S. Duncan
The two basic phenomena that define the problem of visual attention can be illustrated in a simple example and selectivity-the ability to filter out un wanted information is illustrated.
Beyond Bags of Features: Spatial Pyramid Matching for Recognizing Natural Scene Categories
7,929 Citations2006Svetlana Lazebnik, C. Schmid +1 more
This paper presents a method for recognizing scene categories based on approximate global geometric correspondence that exceeds the state of the art on the Caltech-101 database and achieves high accuracy on a large database of fifteen natural scene categories.
IEEE Transactions on Pattern Analysis and Machine IntelligenceA performance evaluation of local descriptors
6,692 Citations2005Krystian Mikolajczyk, C. Schmid
IEEE Transactions on Pattern Analysis and Machine IntelligenceContent-based image retrieval at the end of the early years
6,023 Citations2000A.W.M. Smeulders, Marcel Worring +3 more
The working conditions of content-based retrieval: patterns of use, types of pictures, the role of semantics, and the sensory gap are discussed, as well as aspects of system engineering: databases, system architecture, and evaluation.
Nature NeuroscienceHierarchical models of object recognition in cortex
3,404 Citations1999Maximilian Riesenhuber, Tomaso Poggio
A new hierarchical model consistent with physiological data from inferotemporal cortex that accounts for this complex visual task and makes testable predictions is described.
Lecture notes in computer scienceA Practical Guide to Training Restricted Boltzmann Machines
2,802 Citations2012Geoffrey E. Hinton
This guide is an attempt to share expertise at training restricted Boltzmann machines with other machine learning researchers.
Vlfeat
2,704 Citations2010Andrea Vedaldi, Brian Fulkerson
VLFeat is an open and portable library of computer vision algorithms that includes rigorous implementations of common building blocks such as feature detectors, feature extractors, (hierarchical) k-means clustering, randomized kd-tree matching, and super-pixelization.
Computer Vision and Image UnderstandingLearning generative visual models from few training examples: An incremental Bayesian approach tested on 101 object categories
2,052 Citations2007Li Fei-Fei, Rob Fergus +1 more
The incremental algorithm is compared experimentally to an earlier batch Bayesian algorithm, as well as to one based on maximum-likelihood, which have comparable classification performance on small training sets, but incremental learning is significantly faster, making real-time learning feasible.
The pyramid match kernel: discriminative classification with sets of image features
1,494 Citations2005Kristen Grauman, Trevor Darrell
A new fast kernel function is presented which maps unordered feature sets to multi-resolution histograms and computes a weighted histogram intersection in this space and is shown to be positive-definite, making it valid for use in learning algorithms whose optimal solutions are guaranteed only for Mercer kernels.
Image Classification using Random Forests and Ferns
1,235 Citations2007Anna Bosch, Andrew Zisserman +1 more
It is shown that selecting the ROI adds about 5% to the performance and, together with the other improvements, the result is about a 10% improvement over the state of the art for Caltech-256.
A Theoretical Analysis of Feature Pooling in Visual Recognition
1,148 Citations2010Y-Lan Boureau, Jean Ponce +1 more
It is shown that the reasons underlying the performance of various pooling methods are obscured by several confounding factors, such as the link between the sample cardinality in a spatial pool and the resolution at which low-level features have been extracted.
SVM-KNN: Discriminative Nearest Neighbor Classification for Visual Category Recognition
1,116 Citations2006Hao Zhang, Alexander C. Berg +2 more
This work considers visual category recognition in the framework of measuring similarities, or equivalently perceptual distances, to prototype examples of categories and proposes a hybrid of these two methods which deals naturally with the multiclass setting, has reasonable computational complexity both in training and at run time, and yields excellent results in practice.
In defense of Nearest-Neighbor based image classification
1,072 Citations2008Oren Boiman, Eli Shechtman +1 more
It is argued that two practices commonly used in image classification methods, have led to the inferior performance of NN-based image classifiers: Quantization of local image descriptors (used to generate "bags-of-words ", codebooks) and Computation of 'image-to-image' distance, instead of ' image- to-class' distance.
Journal of NeurophysiologyResponses to Contour Features in Macaque Area V4
553 Citations1999Anitha Pasupathy, Charles E. Connor
The results suggest that V4 processes information about contour features as a step toward complex shape recognition, and a strong bias toward convex features implies a neural basis for the well-known perceptual dominance of convexity.
Journal of NeurophysiologyShape Representation in Area V4: Position-Specific Tuning for Boundary Conformation
494 Citations2001Anitha Pasupathy, Charles E. Connor
Individual V4 cells appear to encode moderately complex boundary information at specific locations within larger shapes, which suggests that, at intermediate stages in the V1-IT transformation, complex objects are represented at least partly in terms of the configurations and positions of their contour components.
Journal of NeurophysiologyNeural responses to polar, hyperbolic, and Cartesian gratings in area V4 of the macaque monkey
454 Citations1996Jack L. Gallant, Charles E. Connor +3 more
The responses of 103 neurons in visual area V4 of anesthetized macaque monkeys to two novel classes of visual stimuli, polar and hyperbolic sinusoidal gratings, were studied to identify cells that had orderly tuning profiles in the various stimulus spaces.
Proceedings of the National Academy of SciencesSymmetry activates extrastriate visual cortex in human and nonhuman primates
252 Citations2005Yuka Sasaki, Wim Vanduffel +3 more
Evidence suggests that visual symmetry is specifically enhanced in the human brain, but that the underlying neural mechanisms may nevertheless be resolvable in nonhuman primates.
Journal of NeurophysiologyA Model of V4 Shape Selectivity and Invariance
236 Citations2007Charles F. Cadieu, Minjoon Kouh +4 more
A quantitative model of hierarchical processing is provided that provides a plausible mechanism for the translation-invariant shape representation observed in area V4 and points toward a possible canonical mechanism operating throughout the ventral pathway.
Journal of NeurophysiologyObject Representations in the Temporal Cortex of Monkeys and Humans as Revealed by Functional Magnetic Resonance Imaging
193 Citations2008Andrew H. Bell, Fadila Hadj‐Bouziane +3 more
FMRI results provide some novel observations and support the idea that representations of complex stimuli in IT cortex are organized into multiple hierarchical tiers, encompassing both semantic and physical properties.
Using Dependent Regions for Object Categorization in a Generative Framework
142 Citations2006Gang Wang, Ye Zhang +1 more
This paper relaxes the independence assumption and model explicitly the inter-dependency of the local regions of the "bag of words" models, and introduces a linkage structure over the latent themes to encode the dependencies of the patches.
Journal of NeurophysiologySpectral Receptive Field Properties Explain Shape Selectivity in Area V4
135 Citations2006Stephen V. David, Benjamin Y. Hayden +1 more
Cortex“Object Vision” and “Spatial Vision”: The Neuropsychological Evidence for the Distinction
94 Citations1990G. Ettlinger
The distinction between "object vision" and "spatial vision" was made by Ungerleider and Mishkin (1982) on the evidence of behavioural and neuroanatomical studies, largely with monkeys, but further work is needed particularly on the spatial disorder after parietal lesions.
Pattern RecognitionLearning dictionary on manifolds for image classification
72 Citations2012Baodi Liu, Yu-Xiong Wang +2 more
The geometrical structures are encoded in two situations, and two topological structures are embedded into sparse coding algorithm as regularization terms to formulate the corresponding objective functions of dictionary learning on single manifold (DLSM) and Dictionary learning on multiple manifolds (DLMM), respectively.
IEEE Transactions on Neural Networks and Learning SystemsLearning and Representing Object Shape Through an Array of Orientation Columns
14 Citations2014Hui Wei, Qiang Li +1 more
The experimental results show that the algorithm can learn from a small training set and can recognize the same types of objects in natural background without any preliminary information, offering a promising prospect for the handling of semantic-concerned problems that need prior knowledge.
Cognitive ComputationBiologically Motivated Model for Outdoor Scene Classification
12 Citations2013Jingjing Zhao, Chun Du +3 more
A novel model that combines biologically inspired features and cortex-like memory patterns is presented that improves the classification accuracy rates to nearly 100 % and significantly reduces training costs compared with other biologically inspired feature-based approaches.
Cognitive ComputationShape Description and Recognition Method Inspired by the Primary Visual Cortex
11 Citations2013Hui Wei, Li Hu
To enable maximum recognition, a moment-based measurement method was designed to describe the layout of active simple cells in each layer in detail and a decision tree was produced using samples re-described using Hu’s moment invariants.
Lecture notes in computer scienceWeighted Pooling Based on Visual Saliency for Image Classification
2 Citations2014Byeongho Heo, Hawook Jeong +3 more
A novel pooling method which decomposes an image into an object region and a background region, and adopts the visual saliency map for a probabilistic measure of objectness to estimate object region from an unlabeled image.