SIFT - scale-invariant feature transform Konrad Schindler

Similar documents
Local Features: Detection, Description & Matching

SIFT: SCALE INVARIANT FEATURE TRANSFORM SURF: SPEEDED UP ROBUST FEATURES BASHAR ALSADIK EOS DEPT. TOPMAP M13 3D GEOINFORMATION FROM IMAGES 2014

Obtaining Feature Correspondences

CS 4495 Computer Vision A. Bobick. CS 4495 Computer Vision. Features 2 SIFT descriptor. Aaron Bobick School of Interactive Computing

SCALE INVARIANT FEATURE TRANSFORM (SIFT)

CAP 5415 Computer Vision Fall 2012

The SIFT (Scale Invariant Feature

Computer Vision for HCI. Topics of This Lecture

Introduction. Introduction. Related Research. SIFT method. SIFT method. Distinctive Image Features from Scale-Invariant. Scale.

Outline 7/2/201011/6/

CEE598 - Visual Sensing for Civil Infrastructure Eng. & Mgmt.

SUMMARY: DISTINCTIVE IMAGE FEATURES FROM SCALE- INVARIANT KEYPOINTS

Local features: detection and description May 12 th, 2015

Scale Invariant Feature Transform

Local features: detection and description. Local invariant features

Local Image Features

Scale Invariant Feature Transform

School of Computing University of Utah

Local Features Tutorial: Nov. 8, 04

Key properties of local features

Feature Detection. Raul Queiroz Feitosa. 3/30/2017 Feature Detection 1

Midterm Wed. Local features: detection and description. Today. Last time. Local features: main components. Goal: interest operator repeatability

Local Image Features

Feature descriptors. Alain Pagani Prof. Didier Stricker. Computer Vision: Object and People Tracking

Local invariant features

Building a Panorama. Matching features. Matching with Features. How do we build a panorama? Computational Photography, 6.882

Motion illusion, rotating snakes

EECS150 - Digital Design Lecture 14 FIFO 2 and SIFT. Recap and Outline

BSB663 Image Processing Pinar Duygulu. Slides are adapted from Selim Aksoy

Local features and image matching. Prof. Xin Yang HUST

Performance Evaluation of Scale-Interpolated Hessian-Laplace and Haar Descriptors for Feature Matching

Feature Based Registration - Image Alignment

Lecture 10 Detectors and descriptors

Motion Estimation and Optical Flow Tracking

EE368 Project Report CD Cover Recognition Using Modified SIFT Algorithm

Ulas Bagci

Local Feature Detectors

2D Image Processing Feature Descriptors

Object Detection by Point Feature Matching using Matlab

Implementing the Scale Invariant Feature Transform(SIFT) Method

Filtering Images. Contents

SIFT: Scale Invariant Feature Transform

CS 558: Computer Vision 4 th Set of Notes

Implementation and Comparison of Feature Detection Methods in Image Mosaicing

Augmented Reality VU. Computer Vision 3D Registration (2) Prof. Vincent Lepetit

A NEW FEATURE BASED IMAGE REGISTRATION ALGORITHM INTRODUCTION

Feature descriptors and matching

Feature Descriptors. CS 510 Lecture #21 April 29 th, 2013

Features Points. Andrea Torsello DAIS Università Ca Foscari via Torino 155, Mestre (VE)

Comparison of Feature Detection and Matching Approaches: SIFT and SURF

Feature Detection and Matching

AK Computer Vision Feature Point Detectors and Descriptors

CS 556: Computer Vision. Lecture 3

Prof. Feng Liu. Spring /26/2017

Edge and Texture. CS 554 Computer Vision Pinar Duygulu Bilkent University

Local Image preprocessing (cont d)

Image Matching. AKA: Image registration, the correspondence problem, Tracking,

3D Reconstruction From Multiple Views Based on Scale-Invariant Feature Transform. Wenqi Zhu

Image Features: Detection, Description, and Matching and their Applications

Coarse-to-fine image registration

3D Object Recognition using Multiclass SVM-KNN

SURF. Lecture6: SURF and HOG. Integral Image. Feature Evaluation with Integral Image

Distinctive Image Features from Scale-Invariant Keypoints

Lecture 7: Most Common Edge Detectors

Lecture 6: Edge Detection

CS 556: Computer Vision. Lecture 3

Distinctive Image Features from Scale-Invariant Keypoints

Computer Vision. Recap: Smoothing with a Gaussian. Recap: Effect of σ on derivatives. Computer Science Tripos Part II. Dr Christopher Town

Ulrik Söderström 16 Feb Image Processing. Segmentation

Faster Image Feature Extraction Hardware

A Novel Algorithm for Color Image matching using Wavelet-SIFT

Patch-based Object Recognition. Basic Idea

Image Features. Work on project 1. All is Vanity, by C. Allan Gilbert,

Scale Invariant Feature Transform by David Lowe

Scale Invariant Feature Transform (SIFT) CS 763 Ajit Rajwade

CS334: Digital Imaging and Multimedia Edges and Contours. Ahmed Elgammal Dept. of Computer Science Rutgers University

Digital Image Processing. Image Enhancement - Filtering

EE795: Computer Vision and Intelligent Systems

Edge and corner detection

Feature Matching and Robust Fitting

Feature-based methods for image matching

Segmentation and Grouping

Vision and Image Processing Lab., CRV Tutorial day- May 30, 2010 Ottawa, Canada

Texture. Frequency Descriptors. Frequency Descriptors. Frequency Descriptors. Frequency Descriptors. Frequency Descriptors

Advanced Video Content Analysis and Video Compression (5LSH0), Module 4

Stitching and Blending

Lecture 4.1 Feature descriptors. Trym Vegard Haavardsholm

Distinctive Image Features from Scale-Invariant Keypoints

Harder case. Image matching. Even harder case. Harder still? by Diva Sian. by swashford

Image Features Detection, Description and Matching

Line, edge, blob and corner detection

Problems with template matching

3D Vision. Viktor Larsson. Spring 2019

Invariant Local Feature for Image Matching

A Comparison of SIFT, PCA-SIFT and SURF

CS534: Introduction to Computer Vision Edges and Contours. Ahmed Elgammal Dept. of Computer Science Rutgers University

Image features. Image Features

EN1610 Image Understanding Lab # 3: Edges

Image Segmentation and Registration

Other Linear Filters CS 211A

Transcription:

SIFT - scale-invariant feature transform Konrad Schindler Institute of Geodesy and Photogrammetry

Invariant interest points Goal match points between images with very different scale, orientation, projective distortion method must be invariant to these transformations (e.g. patches around conventional corner points are not!) Different strategies design an interest point detector, which already compensates for the distortion usual strategy for scale detect the interest point, then normalize the local region around it to a canonical value usual strategy for rotation build invariance to the distortion into the region descriptor usual strategy for other (small) geometric distortions build invariance to the distortion into the similarity measure a possible strategy for photometric changes 2

Image scale-space represent the image signal at all scales (in practice at a discrete number of levels) original image is the highest available scale to create coarser scales, suppress fine details smoothing it can be shown that only Gaussian smoothing ensures that no spurious structures are introduced at coarser scales all levels together represent a Gaussian scale space I σ (x, y)= 1 2πσ e x 2 +y 2 2 2σ 2 I 0 (x, y) I0 =4 =16 =64 3

General strategy run interest point detector on all different scales select strong local maxima in (x,y, )-space Example: Laplace detector blobs of high contrast to the background 4

Regions with strong contrast to their surroundings are local curvature maxima find image locations with maximal 2 nd derivative combine smoothing kernel and Laplace kernel (magnitude of 2 nd derivative) into one filter result: Laplacian-of-Gaussian (LoG) interest point detection by convolution with LoG-kernel 2 G σ = 2 G σ x 2 + 2 G σ y 2

DoG - difference of Gaussians computationally efficient approximation LoG approximated by difference between different scale-space levels LoG = σ 2 G xx (x, y,σ)+g yy (x, y,σ) DoG = G(x, y,kσ) G(x, y,σ) 6

DoG - difference of Gaussians computationally efficient approximation LoG approximated by difference between different scale-space levels 7

Scale selection at which scale is the response maximal for an ideal circle of radius r? zero-crossings of DoG function at circle boundary r signal image Laplacian

scale of interest point = scale with highest DoG response characteristic scale Lindeberg, T., Feature detection with automatic scale selection. Int l Journal of Computer Vision, 1998.

response is also high next to strong gray-value edges edge points are unwanted - cannot be reliably localised only want points where the image intensity has high curvature test value (c.f. Förstner operator) H = 2 Ixx I xy trace(h) I xy I yy det(h) = (I xx +I yy ) 2 I xx I yy I 2 xy <T

Steps create Gaussian scale space compute differences between adjacent levels find maxima in 3-dimensional (x,y, )-space remove responses on edges

Descriptors Goal assign a descriptor to each interest point, which allows to robustly match it across views in aerial photogrammetry: the surrounding image region directly serves as descriptor (e.g. for normalized cross-correlation) problem 1: comparing image regions is not invariant to rotation problem 2: comparing gray-values directly is not invariant to small misalignments due to variations in scale, viewpoint, etc. Solution rotation-invariance (in fact covariance): estimate a dominant rotation, align regions misalignments: use a robust encoding of the region, rather than raw intensity values SIFT descriptor 12

Descriptors Normalizing rotation intuition: we need two find the same direction in object space in the local neighborhoods of corresponding points idea: find the dominant gradient direction in the region 13

Descriptors Encoding local structure raw gradients (or intensities) are sensitive to alignment errors global histograms are invariant to small location errors, but contain no spatial information to be robust against small errors, but preserve some local structure, build histograms over local neighborhoods 0 2 14

Descriptors SIFT descriptor divide patch into 4 x 4 sub-windows compute an 8-bin histogram of gradient orientations each location contributes with its gradient magnitude, contribution is distributed over adjacent bins with trilinear interpolation 15

Descriptors SIFT descriptor result: a 128-dimensional vector (4 x 4 x 8 histogram bins) gradients are invariant to constant offset in brightness invariance to linear contrast scaling: normalize feature vector Lowe, D., Distinctive image features from scale-invariant keypoints, IJCV, 2004. 16

Matching descriptors Similarity measure: Euclidean distance Nearest-neighbor search finding the nearest point in high dimensions is expensive approximate nearest-neighbor search with suitable data structures (kd-trees, vantage-point trees,...) Variant to avoid ambiguities find nearest descriptor v 1 and second-nearest descriptor v 2 check that v 1 is significantly closer than v 2 (say, 30%), otherwise discard match 17