Giessen University Dept. of Psychology

Slides:

Advertisements

Similar presentations

Chapter 2: Marr’s theory of vision. Cognitive Science  José Luis Bermúdez / Cambridge University Press 2010 Overview Introduce Marr’s distinction between.

Advertisements

Midterm 1 Oct. 21 in class. Read this article by Wednesday next week!

A Graph based Geometric Approach to Contour Extraction from Noisy Binary Images Amal Dev Parakkat, Jiju Peethambaran, Philumon Joseph and Ramanathan Muthuganapathy.

Visual Saliency: the signal from V1 to capture attention Li Zhaoping Head, Laboratory of natural intelligence Department of Psychology University College.

The Visual System: Feature Detection Model Lesson 17.

Neural Network Models in Vision Peter Andras

Gabor Filter: A model of visual processing in primary visual cortex (V1) Presented by: CHEN Wei (Rosary) Supervisor: Dr. Richard So.

Example-Based Fractured Appearance L. Glondu, L. Muguercia, M. Marchal, C. Bosch, H. Rushmeier, G. Dumont and G. Drettakis.

Level set based Image Segmentation Hang Xiao Jan12, 2013.

Analysis of Contour Motions Ce Liu William T. Freeman Edward H. Adelson Computer Science and Artificial Intelligence Laboratory Massachusetts Institute.

Computer Vision Lecture 18: Object Recognition II

for image processing and computer vision

A saliency map model explains the effects of random variations along irrelevant dimensions in texture segmentation and visual search Li Zhaoping, University.

Text Detection in Video Min Cai Background  Video OCR: Text detection, extraction and recognition  Detection Target: Artificial text  Text.

A Study of Approaches for Object Recognition

EE663 Image Processing Edge Detection 2 Dr. Samir H. Abdul-Jauwad Electrical Engineering Department King Fahd University of Petroleum & Minerals.

Laurent Itti: CS599 – Computational Architectures in Biological Vision, USC Lecture 6: Low-level features 1 Computational Architectures in Biological.

Brain Theory and Artificial Intelligence

Vision Computing An Introduction. Visual Perception Sight is our most impressive sense. It gives us, without conscious effort, detailed information about.

Segmentation (Section 10.2)

Visual Cognition II Object Perception. Theories of Object Recognition Template matching models Feature matching Models Recognition-by-components Configural.

Laurent Itti: CS599 – Computational Architectures in Biological Vision, USC Lecture 8: Stereoscopic Vision 1 Computational Architectures in Biological.

Michigan State University 1 “Saliency-Based Visual Attention” “Computational Modeling of Visual Attention”, Itti, Koch, (Nature Reviews – Neuroscience.

CS292 Computational Vision and Language Visual Features - Colour and Texture.

Overview 1.The Structure of the Visual Cortex 2.Using Selective Tuning to Model Visual Attention 3.The Motion Hierarchy Model 4.Simulation Results 5.Conclusions.

A neural mechanism for robust junction representation in the visual cortex University of Ulm Dept. of Neural Information Processing Thorsten Hansen and.

Laurent Itti: CS599 – Computational Architectures in Biological Vision, USC Lecture 5: Introduction to Vision 2 1 Computational Architectures in.

West Virginia University

Multi-resolution Arc Segmentation: Algorithms and Performance Evaluation Jiqiang Song Jan. 12 th, 2004.

Studying Visual Attention with the Visual Search Paradigm Marc Pomplun Department of Computer Science University of Massachusetts at Boston

Abstract We start with a statistical formulation of Helmholtz’s ideas about neural energy to furnish a model of perceptual inference and learning that.

The Brain from retina to extrastriate cortex. Neural processing responsible for vision photoreceptors retina –bipolar and horizontal cells –ganglion cells.

Lecture 10: Mean Field theory with fluctuations and correlations Reference: A Lerchner et al, Response Variability in Balanced Cortical Networks, q-bio.NC/ ,

Advances in Modeling Neocortex and its impact on machine intelligence Jeff Hawkins Numenta Inc. VS265 Neural Computation December 2, 2010 Documentation.

Lecture 2b Readings: Kandell Schwartz et al Ch 27 Wolfe et al Chs 3 and 4.

Automatic Minirhizotron Root Image Analysis Using Two-Dimensional Matched Filtering and Local Entropy Thresholding Presented by Guang Zeng.

The primate visual systemHelmuth Radrich, The primate visual system 1.Structure of the eye 2.Neural responses to light 3.Brightness perception.

Supervised Learning of Edges and Object Boundaries Piotr Dollár Zhuowen Tu Serge Belongie.

National Taiwan A Road Sign Recognition System Based on a Dynamic Visual Model C. Y. Fang Department of Information and.

1 Research Question  Can a vision-based mobile robot  with limited computation and memory,  and rapidly varying camera positions,  operate autonomously.

A new theoretical framework for multisensory integration Michael W. Hadley and Elaine R. Reynolds Neuroscience Program, Lafayette College, Easton PA

Complex brain networks: graph theoretical analysis of structural and functional systems.

Colour and Texture. Extract 3-D information Using Vision Extract 3-D information for performing certain tasks such as manipulation, navigation, and recognition.

Visual Computation I. Physiological Foundations

Bayesian inference accounts for the filling-in and suppression of visual perception of bars by context Li Zhaoping 1 & Li Jingling 2 1 University College.

October 1, 2013Computer Vision Lecture 9: From Edges to Contours 1 Canny Edge Detector However, usually there will still be noise in the array E[i, j],

Grenoble Images Parole Signal Automatique Modeling of visual cortical processing to estimate binocular disparity Introduction - The objective is to estimate.

How Recurrent Dynamics Explain Crowding Aaron Clarke & Michael H. Herzog Laboratory of Psychophysics, Brain Mind Institute, École Polytechnique Fédérale.

Machine Vision Edge Detection Techniques ENT 273 Lecture 6 Hema C.R.

A model for detecting illusory contours Caveat: Response decreases gradually with increasing intermeshing of gratings. This does not correspond with subjective.

Computer Vision Image Features Instructor: Dr. Sherif Sami Lecture 4.

Image features and properties. Image content representation The simplest representation of an image pattern is to list image pixels, one after the other.

Ko Youngkil.  Biological Evidence  5 prediction  compare with experimental results  Primary visual cortex (V1) is involved in image segmentation,

Fast Human Detection in Crowded Scenes by Contour Integration and Local Shape Estimation Csaba Beleznai, Horst Bischof Computer Vision and Pattern Recognition,

Activity Recognition Journal Club “Neural Mechanisms for the Recognition of Biological Movements” Martin Giese, Tomaso Poggio (Nature Neuroscience Review,

National Taiwan Normal A System to Detect Complex Motion of Nearby Vehicles on Freeways C. Y. Fang Department of Information.

A Neurodynamical Cortical Model of Visual Attention and Invariant Object Recognition Gustavo Deco Edmund T. Rolls Vision Research, 2004.

Lecture 07 13/12/2011 Shai Avidan הבהרה: החומר המחייב הוא החומר הנלמד בכיתה ולא זה המופיע / לא מופיע במצגת.

By Pradeep C.Venkat Srinath Srinivasan

Brain States: Top-Down Influences in Sensory Processing

Ascending Visual Pathways

Perception: Structures

Bayesian inference Li Zhaoping1 & Li Jingling2 accounts for the

Volume 81, Issue 4, Pages (February 2014)

Caspar M. Schwiedrzik, Winrich A. Freiwald Neuron

Orientation tuning: strongest response to one orientation

Analysis of Contour Motions

The Neural Basis of Perceptual Learning

Volume 75, Issue 1, Pages (July 2012)

Brain States: Top-Down Influences in Sensory Processing

Presentation transcript:

Giessen University Dept. of Psychology Robust contour extraction and junction detection by a neural model utilizing recurrent long-range interactions Thorsten Hansen and Heiko Neumann Giessen University Dept. of Psychology Ulm University Dept. of Neural Information Processing

Overview of the Talk Motivation: empirical evidence for recurrent long-range interactions 2. Approach and Model 3. Results: Contour enhancement Corner detection 4. Conclusions

Sketch of V1 Architecture long-range connections McGuire et al. 1991 LGN recurrent intercolumnar interactions

Specificity of Horizontal Long-Range Connections in V1 Bosking et al. 1997 ”like connects to like” plus colinear arrangement Long-range connections link neurons with same orientation preference and collinear aligned RFs.

Functional Implications of Lateral Long-Range Interactions Polat & Sagi (1993) Measurement of contrast detection thresholds for foveal Gabor elements with and without flankers. Colinear flanking Gabors (up to a distance of 10 wavelengths) facilitate contrast detection.

Key Mechanisms of the Proposed Model Excitatory long-range interactions between cells with collinear aligned RF (Bosking et al. 1997) Inhibitory short-range interactions Modulating feedback: Initial bottom-up activity is necessary (Hirsch & Gilbert 1991)

Model architecture

Recurrent Interaction modulating feedback divisive inhibition inhibition in both spatial and orientational domain excitatory long-range interaction

Results: Contour Enhancement input image complex cells long-range Activity that fits into a more global context is enhanced by top-down feedback.

Results: Temporal Evolution input image complex cells long-range t=1 t =2 t=12

Quantitative Evaluation: Contur Saliency High saliency: large values of (r,z) discrete time steps saliency measures z r input image

Results: Natural Images input image complex cells long-range

Simulation: Physiological Data Kapadia et al. 1995 Simulation: Physiological Data response relative to single bar bar +flankers +texture +flankers+texture enhancement for collinear bar; suppression for noisy textures

Properties of the Proposed Model input image complex cells long-range background: noise suppression corner: preservation of multiple orientations edge: enhancement of coherent structures

Definition of Corners and Junctions Corners and junctions are points where two or more lines join or intersect (from Adelson 2000)

Junctions for Object Recognition (Biederman 1987)

Junctions and Brightness Perception Adelson (2000)

Junction Detection in Natural Images Junctions often cannot be detected locally (McDermott 2001): 13 pixel closeup 13 25 49 97 pixels

Neural Representation of Junctions distributed activity for multiple orientations within a cortical hypercolumn 1. Robust generation of coherent contours model of recurrent long-range interactions in V1 Approach: 2. Read-out of distributed information measure of circular variance

Read-out of Distributed Information Orientation significance: Length of the resulting orientation vector in relation to the overall activity high significance low significance orientation significance circular variance Batschelet 1981: Circular Statistics

Corner and Junction Detection Corner candidates: high circular variance and high overall activity: Corner points: sufficiently large local maxima of corner candidates

Results: Localization Accuracy deviation from true location V1 long-range model feedforward complex cells generic junction configurations

Junction Detection on a Synthetic Image Attneave‘s cat complex cells long-range

Junction Detection on Natural Images Real world camera image complex cells long-range

Junction Detection on Natural Images cut-out of a plant image Van Hateren & van der Schaaf 1998 input image complex cells long-range

Evaluation using ROC Analysis Comparison of the new scheme to standard methods based on Gaussian curvature and the structure tensor (black) input image

Conclusions corners and junctions can be robustly represented by distributed activity within a cortical hypercolum recurrent colinear long-range interactions serve as a multi-purpose mechanism for contour enhancement noise suppression junction detection Hansen & Neumann (2004) Neural Computation 16(5).