efficient color boundary detection with color-opponent mechanisms

26
Efficient Color Boundary Detection with Color- opponent Mechanisms CVPR2013 Posters

Upload: chaela

Post on 05-Jan-2016

48 views

Category:

Documents


0 download

DESCRIPTION

Efficient Color Boundary Detection with Color-opponent Mechanisms. CVPR2013 Posters. Outline. Introduction Approach Experiments Conclusions. Introduction. Introduction. - PowerPoint PPT Presentation

TRANSCRIPT

Page 1: Efficient Color Boundary Detection with Color-opponent Mechanisms

Efficient Color Boundary Detection with Color-opponent Mechanisms

CVPR2013 Posters

Page 2: Efficient Color Boundary Detection with Color-opponent Mechanisms

Outline

Introduction Approach Experiments Conclusions

Page 3: Efficient Color Boundary Detection with Color-opponent Mechanisms

Introduction

Page 4: Efficient Color Boundary Detection with Color-opponent Mechanisms

Introduction

Propose a new framework for boundary detection in complex natural scenes based on the color-opponent mechanisms of the visual system.

Image source:http://en.wikipedia.org/wiki/Opponent_process

Page 5: Efficient Color Boundary Detection with Color-opponent Mechanisms

Introduction One of the key limitations of opponent-based

approaches is that they are

blind to the luminance-defined boundaries. In order to obtain the complete contours of

objects, these methods had to spend extra computational cost to combine more cues to detect luminance boundaries [3].

[3] D. R. Martin, C. C. Fowlkes, and J. Malik, "Learning to detect natural image boundaries using local brightness, color, and texture cues," IEEE Trans. on PAMI, vol. 26, pp. 530-549, 2004.

Page 6: Efficient Color Boundary Detection with Color-opponent Mechanisms

Introduction Simulate the biological mechanisms of

color information processing along the Retina-LGN-Cortex visual pathway

Image source:http://en.wikipedia.org/wiki/Opponent_process

Page 7: Efficient Color Boundary Detection with Color-opponent Mechanisms

Introduction

Image source:[20] S. G. Solomon and P. Lennie, "The machinery of colourvision," Nature Reviews Neuroscience, vol. 8, pp. 276-286, 2007.

Page 8: Efficient Color Boundary Detection with Color-opponent Mechanisms

Introduction

Color Mechanisms in the Visual System. Properties : 1. Trichromacy. 2. Two opponent channels. 3. Color opponency.

Page 9: Efficient Color Boundary Detection with Color-opponent Mechanisms

Approach

Boundary Detection System : 1.Cone Layer 2.Ganglion/LGN Layer 3.Cortex Layer

Page 10: Efficient Color Boundary Detection with Color-opponent Mechanisms

A feedforward hierarchical system

Page 11: Efficient Color Boundary Detection with Color-opponent Mechanisms

1.Cone Layer

Type II cells in the ganglion/LGN layer is mainly for the perception of color region.

Four channels: red (R), green (G), blue (B) and yellow (Y) components, where Y = (R+G)/2.

Gaussian filters are used to simulate the receptive field of the cones in the retina.

Outputs:

Page 12: Efficient Color Boundary Detection with Color-opponent Mechanisms

Approach

Boundary Detection System : 1.Cone Layer 2.Ganglion/LGN Layer 3.Cortex Layer

Page 13: Efficient Color Boundary Detection with Color-opponent Mechanisms

2.Ganglion/LGN Layer

Single-opponent cells in ganglion/LGN layer areimportant for separating color and achromatic information,which is clearly shown by Equation 1.

w1 > 0 and w2 < 0 response : R-on/G-off cellsw1 < 0 and w2 > 0 response : R-off/G-on cells

Page 14: Efficient Color Boundary Detection with Color-opponent Mechanisms

Approach

Boundary Detection System : 1.Cone Layer 2.Ganglion/LGN Layer 3.Cortex Layer

Page 15: Efficient Color Boundary Detection with Color-opponent Mechanisms

3.Cortex Layer

In the cortex layer of V1, the receptive fields of most color- and color-luminance-sensitive neurons are both chromatically and spatially opponent.

Page 16: Efficient Color Boundary Detection with Color-opponent Mechanisms

3.Cortex Layer

Page 17: Efficient Color Boundary Detection with Color-opponent Mechanisms

3.Cortex Layer

The boundary responses at each orientation is given by (6)

Page 18: Efficient Color Boundary Detection with Color-opponent Mechanisms

3.Cortex Layer

The boundaries are detected in four channels (i.e., R+ wG, wR+ G, B+ wY and wB+Y ) with Equations 1-8.

Page 19: Efficient Color Boundary Detection with Color-opponent Mechanisms

Experiments

Page 20: Efficient Color Boundary Detection with Color-opponent Mechanisms

Experiments

Page 21: Efficient Color Boundary Detection with Color-opponent Mechanisms

Experiments

Page 22: Efficient Color Boundary Detection with Color-opponent Mechanisms

Experiments

Page 23: Efficient Color Boundary Detection with Color-opponent Mechanisms

Experiments

Page 24: Efficient Color Boundary Detection with Color-opponent Mechanisms

Experiments

Page 25: Efficient Color Boundary Detection with Color-opponent Mechanisms

Experiments

Page 26: Efficient Color Boundary Detection with Color-opponent Mechanisms

Conclusions

1. Presented a novel biologically plausible computational model for contour detection of color images.

2. Our model exhibits excellent capability of detecting both color and luminance boundaries synchronously in a time-saving manner.