recognition of textures and object classes

Post on 20-Mar-2016

68 Views

Category:

Documents

1 Downloads

Preview:

Click to see full reader

DESCRIPTION

Recognition of textures and object classes. Introduction. Invariant local descriptors => robust recognition of specific objects or scenes Recognition of textures and object classes => description of intra-class variation, selection of discriminant features. texture recognition. - PowerPoint PPT Presentation

TRANSCRIPT

Recognition of

textures and object classes

Introduction

• Invariant local descriptors => robust recognition of specific objects or scenes

• Recognition of textures and object classes => description of intra-class variation, selection of

discriminant features

texture recognition car detection

1. An affine-invariant texture recognition (CVPR’03)

2. A two-layer architecture for texture segmentation and recognition (ICCV’03)

3. Feature selection for object class recognition (ICCV’03)

Overview

Affine-invariant texture recognition

• Texture recognition under viewpoint changes and non-rigid transformations

• Use of affine-invariant regions– invariance to viewpoint changes– spatial selection => more compact representation, reduction of

redundancy in texton dictionary

[A sparse texture representation using affine-invariant regions, S. Lazebnik, C. Schmid and J. Ponce, CVPR 2003]

Overview of the approach

Harris detector

Laplace detector

Region extraction

Descriptors – Spin images

Spatial selection

clustering each pixel

clustering selected pixels

Signature and EMD

• Hierarchical clustering => Signature :

• Earth movers distance

– robust distance, optimizes the flow between distributions– can match signatures of different size– not sensitive to the number of clusters

SS = { ( m1 , w1 ) , … , ( mk , wk ) }

D( SS , SS’’ ) = [i,j fij d( mi , m’j)] / [i,j fij ]

Database with viewpoint changes

20 samples of 10 different textures

Results

Spin images Gabor-like filters

A two-layer architecture

• Texture recognition + segmentation

• Classification of individual regions + spatial layout

[A generative architecture for semi-supervised texture recognition, S. Lazebnik, C. Schmid, J. Ponce, ICCV 2003]

A two-layer architecture

Modeling : 1. Distribution of the local descriptors (affine invariants)

• Gaussian mixture model• estimation with EM, allows incorporating unsegmented images

2. Co-occurrence statistics of sub-class labels over affinely adapted neighborhoods

Segmentation + Recognition :1. Generative model for initial class probabilities2. Co-occurrence statistics + relaxation to improve labels

Texture Dataset – Training Images

T1 (brick) T2 (carpet) T3 (chair) T4 (floor 1) T5 (floor 2) T6 (marble) T7 (wood)

Effect of relaxation + co-occurrence

Original image

Top: before relaxation (indivual regions), bottom: after relaxation (co-occurrence)

Recognition + Segmentation Examples

Animal Dataset – Training Images

• no manual segmentation, weakly supervised• 10 training images per animal (with background) • no purely negative images

Recognition + Segmentation Examples

Object class detection

• Description of intra-class variations of object parts

[Selection of scale inv. regions for object class recognition,G. Dorko and C. Schmid, ICCV’03]

Object class detection

• Description of intra-class variations of object parts

• Selection of discrimiant features

Outline of the approach

Clustering of descriptors

• Descriptors are labeled as positive/negative

• Hierarchical clustering of the positive/negative set

• Examples of positive clusters

Clustering of descriptors

• Descriptors are labeled as positive/negative

• Hierarchical clustering of the positive/negative set

• Examples of positive clusters

Classification

• Learn a separate classifier for each cluster

– Classifier : Support Vector Machine

• Select significant classifiers

– Feature selection with likelihood ratio / mutual information

5

Likelihood Mutual Information

10

25

Likelihood – mutual information

Summary - Approach

• Automatic construction of object part classifiers– scale and rotation invariant– no normalization/alignment of the training and test images

• Selection of discriminant features– interest points, clustering– feature selection with likelihood or mutual information

• Comparison of two feature selection methods– likelihood: more discriminant but very specific– mutual Information: discriminant but not too specific

Material

• Powerpoint presention and papers will be available at

http://www.inrialpes.fr/movi/people/Schmid/cvpr-tutorial03

top related