SIFT, GLOH, SURF descriptors. Dipartimento di Sistemi e Informatica

Size: px
Start display at page:

Download "SIFT, GLOH, SURF descriptors. Dipartimento di Sistemi e Informatica"

Transcription

1 SIFT, GLOH, SURF descriptors Dipartimento di Sistemi e Informatica

2 Invariant local descriptor: Useful for Object RecogniAon and Tracking. Robot LocalizaAon and Mapping. Image RegistraAon and SAtching. Image Retrieval. Augmented Reality (hkp://blogs.oregonstate.edu/hess/sim- library- places- 2nd- in- acm- mm- 10- ossc/) Template Video Stream

3 Scale invariant detectors In most object recogniaon applicaaons, when the scale of the object in the image is unknown instead of extracang features at many different scales and then matching all of them, it is more efficient to design a funcaon on the region which is the same for corresponding regions, even if they are at different scales. The problem can also be stated as follows: given two images of the same scene with a large scale difference between them, find the same interest points independently in each image.

4 For scale invariant feature extracaon it is necessary to detect structures that can be reliably extracted under scale changes. This is done by evaluaang a signature funcaon (a kernel) in the point neighbourhood and plot the result as a funcaon of the neighbourhood scale. Since it measures properaes of the local neighbourhood at a certain scale, it should take a similar qualitaave shape if two keypoints are centered on corresponding image structures; The funcaon shape should be squashed or expanded as a result of the scaling factor. Corresponding neighbourhood sizes should be detected by searching for extrema of the signature funcaon in both images.

5 We can consider points as a funcaon of region size (circle radius). A common approach is to take a local maximum of this funcaon. The soluaon is to search for maxima of suitable funcaons in scale and in space over the images. f Image 1 scale = 1/2 f Image 2 region size region size The region size (scale), for which the maximum is achieved, should be invariant to image scale. f Image 1 f Image 2 scale = 1/2 s 1 region size/scale s 2 region size/scale

6 A good funcaon for scale detecaon has one stable sharp peak f bad f bad f Good! region size region size region size For usual images a good funcaon would be the one with contrast (sharp local intensity change). It is easier to look for zero- crossings of 2 nd derivaave than maxima.

7 There are a few approaches which are truly invariant to significant scale changes. Typically, such techniques assume that the scale change is the same in every direcaon, although they exhibit some robustness to weak affine deformaaons. The appropriate kernel for this is the scale- normalized Gaussian kernel G(x, σ) and its derivaaves.

8 The classical approach is to generate a Gaussian scale- space representaaon of an image, i.e. a set of images from the convoluaon of an isotropic (circular) Gaussian Kernel of various sizes: A larger scale results into a smoother image ExisAng methods search for local extrema in the 3D Gaussian scale- space representaaon of an image (x, y and scale). Local extrema over scale of normalized derivaaves indicate the presence of characterisac local structures The moavaaon for generaang a scale- space representaaon of a given image originates from the basic observaaon that real- world objects are composed of different structures at different scales. This implies that real- world objects, may appear in different ways depending on the scale of observaaon. The Gaussian scale- space guarantees that new structures must not be created when going from a fine scale to any coarser scale. Its properaes include linearity, shim invariance, non- enhancement of local extrema, scale invariance and rotaaonal invariance

9 FuncAons for determining scale Kernels: ( xx(,, ) yy(,, )) 2 L= G x y + G x y σ σ σ Laplacian of Gaussians f = Kernel Image DoG = G( x, y, kσ) G( x, y, σ) Difference of Gaussians (an approximaaon of Laplacian) where Gaussian 2 2 x + y πσ σ Gxy (,, σ ) = e both kernels are invariant to scale and rota8on

10 The method: - build scale- space pyramids; - all scales are examined to idenafy scale- invariant features: - compute the Difference of Gaussian (DoG) pyramid or Laplacian of Gaussians (LoG) - detect maxima and minima in scale space scale Harris- Laplacian 1 Find local maximum of: Harris corner detector in space (image coordinates) Laplacian in scale y Laplacian Harris x SIFT (Lowe) 2 Find local maximum of: Difference of Gaussians in space and scale scale y DoG DoG 1 K.Mikolajczyk, C.Schmid. Indexing Based on Scale Invariant Interest Points. ICCV D.Lowe. DisAncAve Image Features from Scale- Invariant Keypoints. Accepted to IJCV 2004 x

11 Harris- Laplacian scale- invariant detector Harris- Laplacian method uses Harris funcaon first at mulaple scales, then selects points for which Laplacian akains maximum over scales. Harris- corner points are interest points that have good rotaaonal and illuminaaon invariance. But are not scale invariant. To reflect scale- invariance the second- moment matrix is modified taking a Gaussian scale space representaaon with a Laplacian of Gaussian kernel. Since the computaaon of derivaaves usually involves a stage of scale- space smoothing, an operaaonal definiaon of the Harris operator requires two scale parameters: (i) a local deriva8on scale for smoothing before the computaaon of derivaaves (ii) an integra8on scale for accumulaang the operaaons on derivaaves where g(σ I ) is the Gaussian kernel of scale σ I (integraaon scale) and L(x,y) is the gaussian smoothed image and L x and L y its derivaaves in the x and y direcaon, calculated using a Gaussian kernel of scale σ D (differenaaaon scale). MulAplicaAon by σ 2 is because derivaaves must be normalized across scales according to D m (x, s ) = σ m L m (x, s ).

12 The algorithm searches across mulaple scales σ n σ 0, k 1 σ 0, k 2 σ 0 k 3 σ D k n σ 0 (k=1,4 ) sekng σ I = σ n and σ D = s σ I (s=0,7). At each scale corners are found as with the Harris method applied to M matrix in a 8 point neighbourhood. An iteraave algorithm localizes corner points spaaally and chooses the characterisac scale: Laplacian of Gaussians is used to judge if each of the candidate points found on different levels, forms a maximum in the scale direcaon (check with n- 1 and n+1). The scale where such maximum in scale is found is referred to as CharacterisAc scale. It is used in future iteraaons. Points are spaaally localized at the characterisac scale Mikolajczyk and Schmid (2001) demonstrated that the LoG measure D D D akains the highest percentage of correctly detected corner points in comparison to other scale- selecaon measures: At each iteraaon the corner point x k+1 is selected that maximizes the LoG within the scale neigbourhood. The process terminates when x k+1 = x k

13 MulA- scale Harris points SelecAon of points at the characterisac scale with Laplacian Invariant points + associated regions [Mikolajczyk & Schmid 01]

14 SIFT Scale Invariant Feature Transform SIFT method has been introduced by D. Lowe in 2004 to represent visual enaaes according to their local properaes. The method employs local features taken in correspondence of salient points (referred to as keypoints or SIFT points). Keypoints (their SIFT descriptors) are used to characterize shapes with invariant properaes Image points selected as keypoints and their SIFT descriptors are robust under: - Luminance change (due to difference- based metrics) - Scale change (due to scale- space) RotaAon (due to local orientaaons wrt the keypoint canonical) - The original Lowe s algorithm: Given a grey- scale image: - Build a Gaussian- blurred image pyramid - Subtract adjacent levels to obtain a Difference of Gaussians (DoG) pyramid (so approximaang the Laplacian of Gaussians) - Take local extrema of the DoG filters at different scales as keypoints - Compute keypoint dominant orientaaon For each keypoint: - Evaluate local gradients in a neighbourhood of the keypoint with orientaaons relaave to the keypoint orientaaon and normalize Build a descriptor as a feature vector with the salient keypoint informaaon -

15 MoAvaAons for usage of DoG are that while Laplacian of Gaussian σ 2 2 G (x,y, σ) provides strong responses to dark blobs of size σ and is good to capture scale invariance, calculaaon of Laplacian is costly. So an approximaaon can be used: Scale normalized Laplacian σ 2 G (x,y, σ) Heat diffusion equaaon unless ½ mulaplicaave constant (Koenderink 92 for luminance scale space) SIFT descriptors are obtained in the following three steps: 1. Keypoint detecaon using local extrema of DoG filters 2. ComputaAon of keypoint orientaaon 3. SIFT descriptor derivaaon

16 Build Gaussian pyramids Keypoints are detected as local scale- space maxima of the Differences of Gaussians. They correspond to local min/max points in image I(x,y) that keep stable at different scales σ Resample Blur Pyramid construcaon process Blur: σ is doubled from the bokom to top of each pyramid Resample: pyramid images are sub- sampled from scale to scale Subtract: adjacent levels of pyramid images are subtracted

17 Building pyramids in detail A first pyramid is obtained by the convoluaon operaaon at different σ such that σ n =k n σ 0 L(x,y,σ) are grouped into a first octave The DoG at a scale σ is obtained by the difference of two nearby scales separated by a constant k AMer the first octave is completed the image such that σ = 2 σ 0 is subsampled by a factor equal to 2 and the next pyramid is obtained in the same way The procedure is iterated for the next levels σ n =k n σ 0 L(x, y, σ) = G(x, y, σ) *I(x, y) D(x, y, σ) = L(x, y, kσ) L(x, y, σ) σ 0 = (k ) 0 σ σ 1 = (k ) 1 σ σ 2 = (k ) 2 σ σ 3 = (k ) 3 σ σ 4 = (k ) 4 σ

18 Octave: the original image is convoluted with a set of Gaussians, so as to obtain a set of images that differ by k in the scale space: each of these sets is usually called octave. k 4 k 3 Octave k 2 k k 0 Each octave is divided into a number of intervals such as k = 2 1/s.. For each octave s + 3 images must be calculated. For example if s = 2 then k = 2 ½ and we will have 5 images at different scales. In this case an octave corresponds to doubling the value of σ σ 0 = (2 ½ ) 0 σ = σ σ 1 = (2 ½ ) 1 σ = κ σ σ 2 = (2 ½ ) 2 σ = 2 σ σ 3 = (2 ½ ) 3 σ = 2 κ σ σ 4 = (2 ½ ) 4 σ = κ 2 σ σ 4 is doubled wrt σ 0

19 Choice of s = 2 is based on empirical verificaaon of the keypoint stability

20 Gaussian kernel size: the number of samples increases as σ increases. The number of operaaons that are needed are (N 2-1) sums and N 2 products. They grow as σ grows. A good compromise is to use a sample interval of [-3σ, 3σ] Sums (N 2 1) Products (N 2 )

21 ComputaAonal savings can be obtained considering that the Gaussian kernel is separable into the product of two one- dimensional convoluaons (2N) products and (2N 2) sums. This makes computaaonal complexity O(N). Moreover convoluaon of two gaussians of σ 1 2 and σ 2 2 is a Gaussian with variance: σ 3 2 = ( σ σ 2 2 ). This property can be exploited to build the scale space, so to use convoluaons already calculated

22 aa Example

23 Detect maxima and minima of DoGs in scale space Local extrema of D(x,y,σ) are the local interest points. To detect the interest points at each level of scale of the DoG pyramid every pixel p is compared to its 8 neighbours: if p is a local extrema (local minimum or maximum) it is selected as a candidate keypoint each candidate keypoint is compared to the 9 neighbours in the scale above and below Only pixels that are local extrema in 3 adjacent levels are promoted as keypoints

24 Keypoint stability The many points extracted from maxima+minima of DoGs have only pixel- accuracy at best and may correspond to low contrast and therefore unreliable points. To improve keypoint stability a funcaon is adapted to the local points in order to determine the interpolated posiaon. Since points are defined in 3D (x,y, σ) it is a 3D curve fikng problem. The interpolaaon is done using the quadraac Taylor expansion of the Difference- of- Gaussian scale- space funcaon, with the candidate keypoint as the origin: k x w- k where D and its derivaaves are evaluated at the candidate keypoint k (x,y σ) and x(x,y σ) is the offset from this point. The locaaon of the extremum, is determined by taking the derivaave of this funcaon with respect to x and sekng it to zero: that is where

25 If the offset is larger than 0.5 in any dimension, then it is an indicaaon that the extremum lies closer to another candidate keypoint. In this case, the candidate keypoint is changed and the interpolaaon performed instead about that point. otherwise the offset is added to its candidate keypoint to get the interpolated esamate for the locaaon of the extremum.

26 low contrast keypoints are generally less reliable than high contrast and keypoints that respond to edges are unstable. Filtering can be performed respecavely by: thresholding on simple contrast thresholding based on principal curvature The local contrast can be directly obtained from D(x,y,σ) calculated at the locaaon of the keypoint as updated from the previous step. Unstable extrema with low contrast can be discarded according to Lowe rule: D(x) < 0,03

27 The DoG funcaon has strong responses along edges. To eliminate the keypoints that have poorly determined locaaons but have high edge responses it must be noaced that for poorly defined peaks in the DoG funcaon, the principal curvature across the edge would be much larger than the principal curvature along it. Finding these principal curvatures amounts to solving for the eigenvalues of the second- order Hessian matrix of D(x,y,s). The eigenvalues of H are proporaonal to the principal curvature of D(x,y,s): to calculate for adiacent DoG pixels The raao of the two eigenvalues is sufficient to the goal. If r is the raao between the highest and the lowest eigenvalue, then: R = (r+1) 2 / r depends only on the raao of the two eigenvalues and is minimum when the two eigenvalues have the same value and increases as r increases. In order to have the raao between the two principal curvatures below a threshold it must be that for some threshold on r, if R is higher than the keypoint is poorly localized and hence rejected.

28 Maxima in D Remove low contrast and edges

29 Experimental evaluaaon of detectors w.r.t. scale change Repeatability rate: # correspondences # possible correspondences (points present)

30 The common drawback of both the LoG (and DoG) representaaon is that local maxima can also be detected in the neighborhood of contours or straight edges, where the signal change is only in one direcaon. These maxima are less stable because their localizaaon is more sensiave to noise or small changes in neighboring texture.

31 OrientaAon assignment For a keypoint, if L is the image with the closest scale, for a region around keypoint compute gradient magnitude and orientaaon using finite differences: GradientVector Lx ( + 1, y) Lx ( 1, y) = Lxy (, + 1) Lxy (, 1)

32 For such region - create an histogram with 36 bins for orientaaon - weight each point with Gaussian window of 1.5σ east squares)

33 Peak orientaaon is the keypoint canonical orientaaon Any peak within 80% of the highest peak is used to create a keypoint with that orientaaon. Local peak within 80% creates mulaple orientaaons. About 15% has mulaple orientaaons

34 Once the local orientaaon and scale of a keypoint have been esamated, a scaled and oriented patch around the detected point can be extracted and used to form a feature descriptor

INTEREST POINTS AT DIFFERENT SCALES

INTEREST POINTS AT DIFFERENT SCALES INTEREST POINTS AT DIFFERENT SCALES Thank you for the slides. They come mostly from the following sources. Dan Huttenlocher Cornell U David Lowe U. of British Columbia Martial Hebert CMU Intuitively, junctions

More information

LoG Blob Finding and Scale. Scale Selection. Blobs (and scale selection) Achieving scale covariance. Blob detection in 2D. Blob detection in 2D

LoG Blob Finding and Scale. Scale Selection. Blobs (and scale selection) Achieving scale covariance. Blob detection in 2D. Blob detection in 2D Achieving scale covariance Blobs (and scale selection) Goal: independently detect corresponding regions in scaled versions of the same image Need scale selection mechanism for finding characteristic region

More information

Lecture 8: Interest Point Detection. Saad J Bedros

Lecture 8: Interest Point Detection. Saad J Bedros #1 Lecture 8: Interest Point Detection Saad J Bedros sbedros@umn.edu Review of Edge Detectors #2 Today s Lecture Interest Points Detection What do we mean with Interest Point Detection in an Image Goal:

More information

Achieving scale covariance

Achieving scale covariance Achieving scale covariance Goal: independently detect corresponding regions in scaled versions of the same image Need scale selection mechanism for finding characteristic region size that is covariant

More information

Blobs & Scale Invariance

Blobs & Scale Invariance Blobs & Scale Invariance Prof. Didier Stricker Doz. Gabriele Bleser Computer Vision: Object and People Tracking With slides from Bebis, S. Lazebnik & S. Seitz, D. Lowe, A. Efros 1 Apertizer: some videos

More information

Vlad Estivill-Castro (2016) Robots for People --- A project for intelligent integrated systems

Vlad Estivill-Castro (2016) Robots for People --- A project for intelligent integrated systems 1 Vlad Estivill-Castro (2016) Robots for People --- A project for intelligent integrated systems V. Estivill-Castro 2 Perception Concepts Vision Chapter 4 (textbook) Sections 4.3 to 4.5 What is the course

More information

CSE 473/573 Computer Vision and Image Processing (CVIP)

CSE 473/573 Computer Vision and Image Processing (CVIP) CSE 473/573 Computer Vision and Image Processing (CVIP) Ifeoma Nwogu inwogu@buffalo.edu Lecture 11 Local Features 1 Schedule Last class We started local features Today More on local features Readings for

More information

Scale-space image processing

Scale-space image processing Scale-space image processing Corresponding image features can appear at different scales Like shift-invariance, scale-invariance of image processing algorithms is often desirable. Scale-space representation

More information

Feature detectors and descriptors. Fei-Fei Li

Feature detectors and descriptors. Fei-Fei Li Feature detectors and descriptors Fei-Fei Li Feature Detection e.g. DoG detected points (~300) coordinates, neighbourhoods Feature Description e.g. SIFT local descriptors (invariant) vectors database of

More information

Advances in Computer Vision. Prof. Bill Freeman. Image and shape descriptors. Readings: Mikolajczyk and Schmid; Belongie et al.

Advances in Computer Vision. Prof. Bill Freeman. Image and shape descriptors. Readings: Mikolajczyk and Schmid; Belongie et al. 6.869 Advances in Computer Vision Prof. Bill Freeman March 3, 2005 Image and shape descriptors Affine invariant features Comparison of feature descriptors Shape context Readings: Mikolajczyk and Schmid;

More information

SIFT: SCALE INVARIANT FEATURE TRANSFORM BY DAVID LOWE

SIFT: SCALE INVARIANT FEATURE TRANSFORM BY DAVID LOWE SIFT: SCALE INVARIANT FEATURE TRANSFORM BY DAVID LOWE Overview Motivation of Work Overview of Algorithm Scale Space and Difference of Gaussian Keypoint Localization Orientation Assignment Descriptor Building

More information

Lecture 7: Finding Features (part 2/2)

Lecture 7: Finding Features (part 2/2) Lecture 7: Finding Features (part 2/2) Professor Fei- Fei Li Stanford Vision Lab Lecture 7 -! 1 What we will learn today? Local invariant features MoHvaHon Requirements, invariances Keypoint localizahon

More information

Feature detectors and descriptors. Fei-Fei Li

Feature detectors and descriptors. Fei-Fei Li Feature detectors and descriptors Fei-Fei Li Feature Detection e.g. DoG detected points (~300) coordinates, neighbourhoods Feature Description e.g. SIFT local descriptors (invariant) vectors database of

More information

Lecture 12. Local Feature Detection. Matching with Invariant Features. Why extract features? Why extract features? Why extract features?

Lecture 12. Local Feature Detection. Matching with Invariant Features. Why extract features? Why extract features? Why extract features? Lecture 1 Why extract eatures? Motivation: panorama stitching We have two images how do we combine them? Local Feature Detection Guest lecturer: Alex Berg Reading: Harris and Stephens David Lowe IJCV We

More information

Corners, Blobs & Descriptors. With slides from S. Lazebnik & S. Seitz, D. Lowe, A. Efros

Corners, Blobs & Descriptors. With slides from S. Lazebnik & S. Seitz, D. Lowe, A. Efros Corners, Blobs & Descriptors With slides from S. Lazebnik & S. Seitz, D. Lowe, A. Efros Motivation: Build a Panorama M. Brown and D. G. Lowe. Recognising Panoramas. ICCV 2003 How do we build panorama?

More information

Detectors part II Descriptors

Detectors part II Descriptors EECS 442 Computer vision Detectors part II Descriptors Blob detectors Invariance Descriptors Some slides of this lectures are courtesy of prof F. Li, prof S. Lazebnik, and various other lecturers Goal:

More information

Overview. Introduction to local features. Harris interest points + SSD, ZNCC, SIFT. Evaluation and comparison of different detectors

Overview. Introduction to local features. Harris interest points + SSD, ZNCC, SIFT. Evaluation and comparison of different detectors Overview Introduction to local features Harris interest points + SSD, ZNCC, SIFT Scale & affine invariant interest point detectors Evaluation and comparison of different detectors Region descriptors and

More information

Edges and Scale. Image Features. Detecting edges. Origin of Edges. Solution: smooth first. Effects of noise

Edges and Scale. Image Features. Detecting edges. Origin of Edges. Solution: smooth first. Effects of noise Edges and Scale Image Features From Sandlot Science Slides revised from S. Seitz, R. Szeliski, S. Lazebnik, etc. Origin of Edges surface normal discontinuity depth discontinuity surface color discontinuity

More information

Extract useful building blocks: blobs. the same image like for the corners

Extract useful building blocks: blobs. the same image like for the corners Extract useful building blocks: blobs the same image like for the corners Here were the corners... Blob detection in 2D Laplacian of Gaussian: Circularly symmetric operator for blob detection in 2D 2 g=

More information

Lecture 7: Finding Features (part 2/2)

Lecture 7: Finding Features (part 2/2) Lecture 7: Finding Features (part 2/2) Dr. Juan Carlos Niebles Stanford AI Lab Professor Fei- Fei Li Stanford Vision Lab 1 What we will learn today? Local invariant features MoPvaPon Requirements, invariances

More information

Invariant local features. Invariant Local Features. Classes of transformations. (Good) invariant local features. Case study: panorama stitching

Invariant local features. Invariant Local Features. Classes of transformations. (Good) invariant local features. Case study: panorama stitching Invariant local eatures Invariant Local Features Tuesday, February 6 Subset o local eature types designed to be invariant to Scale Translation Rotation Aine transormations Illumination 1) Detect distinctive

More information

Feature extraction: Corners and blobs

Feature extraction: Corners and blobs Feature extraction: Corners and blobs Review: Linear filtering and edge detection Name two different kinds of image noise Name a non-linear smoothing filter What advantages does median filtering have over

More information

SIFT keypoint detection. D. Lowe, Distinctive image features from scale-invariant keypoints, IJCV 60 (2), pp , 2004.

SIFT keypoint detection. D. Lowe, Distinctive image features from scale-invariant keypoints, IJCV 60 (2), pp , 2004. SIFT keypoint detection D. Lowe, Distinctive image features from scale-invariant keypoints, IJCV 60 (), pp. 91-110, 004. Keypoint detection with scale selection We want to extract keypoints with characteristic

More information

Corner detection: the basic idea

Corner detection: the basic idea Corner detection: the basic idea At a corner, shifting a window in any direction should give a large change in intensity flat region: no change in all directions edge : no change along the edge direction

More information

Local Features (contd.)

Local Features (contd.) Motivation Local Features (contd.) Readings: Mikolajczyk and Schmid; F&P Ch 10 Feature points are used also or: Image alignment (homography, undamental matrix) 3D reconstruction Motion tracking Object

More information

Overview. Harris interest points. Comparing interest points (SSD, ZNCC, SIFT) Scale & affine invariant interest points

Overview. Harris interest points. Comparing interest points (SSD, ZNCC, SIFT) Scale & affine invariant interest points Overview Harris interest points Comparing interest points (SSD, ZNCC, SIFT) Scale & affine invariant interest points Evaluation and comparison of different detectors Region descriptors and their performance

More information

Image Analysis. Feature extraction: corners and blobs

Image Analysis. Feature extraction: corners and blobs Image Analysis Feature extraction: corners and blobs Christophoros Nikou cnikou@cs.uoi.gr Images taken from: Computer Vision course by Svetlana Lazebnik, University of North Carolina at Chapel Hill (http://www.cs.unc.edu/~lazebnik/spring10/).

More information

Blob Detection CSC 767

Blob Detection CSC 767 Blob Detection CSC 767 Blob detection Slides: S. Lazebnik Feature detection with scale selection We want to extract features with characteristic scale that is covariant with the image transformation Blob

More information

SURF Features. Jacky Baltes Dept. of Computer Science University of Manitoba WWW:

SURF Features. Jacky Baltes Dept. of Computer Science University of Manitoba   WWW: SURF Features Jacky Baltes Dept. of Computer Science University of Manitoba Email: jacky@cs.umanitoba.ca WWW: http://www.cs.umanitoba.ca/~jacky Salient Spatial Features Trying to find interest points Points

More information

Harris Corner Detector

Harris Corner Detector Multimedia Computing: Algorithms, Systems, and Applications: Feature Extraction By Dr. Yu Cao Department of Computer Science The University of Massachusetts Lowell Lowell, MA 01854, USA Part of the slides

More information

Lecture 8: Interest Point Detection. Saad J Bedros

Lecture 8: Interest Point Detection. Saad J Bedros #1 Lecture 8: Interest Point Detection Saad J Bedros sbedros@umn.edu Last Lecture : Edge Detection Preprocessing of image is desired to eliminate or at least minimize noise effects There is always tradeoff

More information

SIFT: Scale Invariant Feature Transform

SIFT: Scale Invariant Feature Transform 1 SIFT: Scale Invariant Feature Transform With slides from Sebastian Thrun Stanford CS223B Computer Vision, Winter 2006 3 Pattern Recognition Want to find in here SIFT Invariances: Scaling Rotation Illumination

More information

CS 3710: Visual Recognition Describing Images with Features. Adriana Kovashka Department of Computer Science January 8, 2015

CS 3710: Visual Recognition Describing Images with Features. Adriana Kovashka Department of Computer Science January 8, 2015 CS 3710: Visual Recognition Describing Images with Features Adriana Kovashka Department of Computer Science January 8, 2015 Plan for Today Presentation assignments + schedule changes Image filtering Feature

More information

Wavelet-based Salient Points with Scale Information for Classification

Wavelet-based Salient Points with Scale Information for Classification Wavelet-based Salient Points with Scale Information for Classification Alexandra Teynor and Hans Burkhardt Department of Computer Science, Albert-Ludwigs-Universität Freiburg, Germany {teynor, Hans.Burkhardt}@informatik.uni-freiburg.de

More information

Edge Detection. CS 650: Computer Vision

Edge Detection. CS 650: Computer Vision CS 650: Computer Vision Edges and Gradients Edge: local indication of an object transition Edge detection: local operators that find edges (usually involves convolution) Local intensity transitions are

More information

Lecture 7: Edge Detection

Lecture 7: Edge Detection #1 Lecture 7: Edge Detection Saad J Bedros sbedros@umn.edu Review From Last Lecture Definition of an Edge First Order Derivative Approximation as Edge Detector #2 This Lecture Examples of Edge Detection

More information

Scale & Affine Invariant Interest Point Detectors

Scale & Affine Invariant Interest Point Detectors Scale & Affine Invariant Interest Point Detectors KRYSTIAN MIKOLAJCZYK AND CORDELIA SCHMID [2004] Shreyas Saxena Gurkirit Singh 23/11/2012 Introduction We are interested in finding interest points. What

More information

Advanced Features. Advanced Features: Topics. Jana Kosecka. Slides from: S. Thurn, D. Lowe, Forsyth and Ponce. Advanced features and feature matching

Advanced Features. Advanced Features: Topics. Jana Kosecka. Slides from: S. Thurn, D. Lowe, Forsyth and Ponce. Advanced features and feature matching Advanced Features Jana Kosecka Slides from: S. Thurn, D. Lowe, Forsyth and Ponce Advanced Features: Topics Advanced features and feature matching Template matching SIFT features Haar features 2 1 Features

More information

CS5670: Computer Vision

CS5670: Computer Vision CS5670: Computer Vision Noah Snavely Lecture 5: Feature descriptors and matching Szeliski: 4.1 Reading Announcements Project 1 Artifacts due tomorrow, Friday 2/17, at 11:59pm Project 2 will be released

More information

EE 6882 Visual Search Engine

EE 6882 Visual Search Engine EE 6882 Visual Search Engine Prof. Shih Fu Chang, Feb. 13 th 2012 Lecture #4 Local Feature Matching Bag of Word image representation: coding and pooling (Many slides from A. Efors, W. Freeman, C. Kambhamettu,

More information

Feature detection.

Feature detection. Feature detection Kim Steenstrup Pedersen kimstp@itu.dk The IT University of Copenhagen Feature detection, The IT University of Copenhagen p.1/20 What is a feature? Features can be thought of as symbolic

More information

Properties of detectors Edge detectors Harris DoG Properties of descriptors SIFT HOG Shape context

Properties of detectors Edge detectors Harris DoG Properties of descriptors SIFT HOG Shape context Lecture 10 Detectors and descriptors Properties of detectors Edge detectors Harris DoG Properties of descriptors SIFT HOG Shape context Silvio Savarese Lecture 10-16-Feb-15 From the 3D to 2D & vice versa

More information

Overview. Introduction to local features. Harris interest points + SSD, ZNCC, SIFT. Evaluation and comparison of different detectors

Overview. Introduction to local features. Harris interest points + SSD, ZNCC, SIFT. Evaluation and comparison of different detectors Overview Introduction to local features Harris interest points + SSD, ZNCC, SIFT Scale & affine invariant interest point detectors Evaluation and comparison of different detectors Region descriptors and

More information

Image Processing 1 (IP1) Bildverarbeitung 1

Image Processing 1 (IP1) Bildverarbeitung 1 MIN-Fakultät Fachbereich Informatik Arbeitsbereich SAV/BV KOGS Image Processing 1 IP1 Bildverarbeitung 1 Lecture : Object Recognition Winter Semester 015/16 Slides: Prof. Bernd Neumann Slightly revised

More information

Recap: edge detection. Source: D. Lowe, L. Fei-Fei

Recap: edge detection. Source: D. Lowe, L. Fei-Fei Recap: edge detection Source: D. Lowe, L. Fei-Fei Canny edge detector 1. Filter image with x, y derivatives of Gaussian 2. Find magnitude and orientation of gradient 3. Non-maximum suppression: Thin multi-pixel

More information

EECS150 - Digital Design Lecture 15 SIFT2 + FSM. Recap and Outline

EECS150 - Digital Design Lecture 15 SIFT2 + FSM. Recap and Outline EECS150 - Digital Design Lecture 15 SIFT2 + FSM Oct. 15, 2013 Prof. Ronald Fearing Electrical Engineering and Computer Sciences University of California, Berkeley (slides courtesy of Prof. John Wawrzynek)

More information

Edge Detection PSY 5018H: Math Models Hum Behavior, Prof. Paul Schrater, Spring 2005

Edge Detection PSY 5018H: Math Models Hum Behavior, Prof. Paul Schrater, Spring 2005 Edge Detection PSY 5018H: Math Models Hum Behavior, Prof. Paul Schrater, Spring 2005 Gradients and edges Points of sharp change in an image are interesting: change in reflectance change in object change

More information

Filtering and Edge Detection

Filtering and Edge Detection Filtering and Edge Detection Local Neighborhoods Hard to tell anything from a single pixel Example: you see a reddish pixel. Is this the object s color? Illumination? Noise? The next step in order of complexity

More information

Feature Vector Similarity Based on Local Structure

Feature Vector Similarity Based on Local Structure Feature Vector Similarity Based on Local Structure Evgeniya Balmachnova, Luc Florack, and Bart ter Haar Romeny Eindhoven University of Technology, P.O. Box 53, 5600 MB Eindhoven, The Netherlands {E.Balmachnova,L.M.J.Florack,B.M.terHaarRomeny}@tue.nl

More information

Gaussian derivatives

Gaussian derivatives Gaussian derivatives UCU Winter School 2017 James Pritts Czech Tecnical University January 16, 2017 1 Images taken from Noah Snavely s and Robert Collins s course notes Definition An image (grayscale)

More information

Edge Detection in Computer Vision Systems

Edge Detection in Computer Vision Systems 1 CS332 Visual Processing in Computer and Biological Vision Systems Edge Detection in Computer Vision Systems This handout summarizes much of the material on the detection and description of intensity

More information

Lesson 04. KAZE, Non-linear diffusion filtering, ORB, MSER. Ing. Marek Hrúz, Ph.D.

Lesson 04. KAZE, Non-linear diffusion filtering, ORB, MSER. Ing. Marek Hrúz, Ph.D. Lesson 04 KAZE, Non-linear diffusion filtering, ORB, MSER Ing. Marek Hrúz, Ph.D. Katedra Kybernetiky Fakulta aplikovaných věd Západočeská univerzita v Plzni Lesson 04 KAZE ORB: an efficient alternative

More information

Scale & Affine Invariant Interest Point Detectors

Scale & Affine Invariant Interest Point Detectors Scale & Affine Invariant Interest Point Detectors Krystian Mikolajczyk and Cordelia Schmid Presented by Hunter Brown & Gaurav Pandey, February 19, 2009 Roadmap: Motivation Scale Invariant Detector Affine

More information

Instance-level recognition: Local invariant features. Cordelia Schmid INRIA, Grenoble

Instance-level recognition: Local invariant features. Cordelia Schmid INRIA, Grenoble nstance-level recognition: ocal invariant features Cordelia Schmid NRA Grenoble Overview ntroduction to local features Harris interest points + SSD ZNCC SFT Scale & affine invariant interest point detectors

More information

STA 414/2104: Machine Learning

STA 414/2104: Machine Learning STA 414/2104: Machine Learning Russ Salakhutdinov Department of Computer Science! Department of Statistics! rsalakhu@cs.toronto.edu! h0p://www.cs.toronto.edu/~rsalakhu/ Lecture 2 Linear Least Squares From

More information

Lecture 6: Edge Detection. CAP 5415: Computer Vision Fall 2008

Lecture 6: Edge Detection. CAP 5415: Computer Vision Fall 2008 Lecture 6: Edge Detection CAP 5415: Computer Vision Fall 2008 Announcements PS 2 is available Please read it by Thursday During Thursday lecture, I will be going over it in some detail Monday - Computer

More information

Maximum Likelihood (ML), Expecta6on Maximiza6on (EM)

Maximum Likelihood (ML), Expecta6on Maximiza6on (EM) Maximum Likelihood (ML), Expecta6on Maximiza6on (EM) Pieter Abbeel UC Berkeley EECS Many slides adapted from Thrun, Burgard and Fox, ProbabilisAc RoboAcs Outline Maximum likelihood (ML) Priors, and maximum

More information

Affine Adaptation of Local Image Features Using the Hessian Matrix

Affine Adaptation of Local Image Features Using the Hessian Matrix 29 Advanced Video and Signal Based Surveillance Affine Adaptation of Local Image Features Using the Hessian Matrix Ruan Lakemond, Clinton Fookes, Sridha Sridharan Image and Video Research Laboratory Queensland

More information

Maximally Stable Local Description for Scale Selection

Maximally Stable Local Description for Scale Selection Maximally Stable Local Description for Scale Selection Gyuri Dorkó and Cordelia Schmid INRIA Rhône-Alpes, 655 Avenue de l Europe, 38334 Montbonnot, France {gyuri.dorko,cordelia.schmid}@inrialpes.fr Abstract.

More information

Laplacian Filters. Sobel Filters. Laplacian Filters. Laplacian Filters. Laplacian Filters. Laplacian Filters

Laplacian Filters. Sobel Filters. Laplacian Filters. Laplacian Filters. Laplacian Filters. Laplacian Filters Sobel Filters Note that smoothing the image before applying a Sobel filter typically gives better results. Even thresholding the Sobel filtered image cannot usually create precise, i.e., -pixel wide, edges.

More information

Linear Diffusion. E9 242 STIP- R. Venkatesh Babu IISc

Linear Diffusion. E9 242 STIP- R. Venkatesh Babu IISc Linear Diffusion Derivation of Heat equation Consider a 2D hot plate with Initial temperature profile I 0 (x, y) Uniform (isotropic) conduction coefficient c Unit thickness (along z) Problem: What is temperature

More information

Edge Detection. Introduction to Computer Vision. Useful Mathematics Funcs. The bad news

Edge Detection. Introduction to Computer Vision. Useful Mathematics Funcs. The bad news Edge Detection Introduction to Computer Vision CS / ECE 8B Thursday, April, 004 Edge detection (HO #5) Edge detection is a local area operator that seeks to find significant, meaningful changes in image

More information

Lecture 6: Finding Features (part 1/2)

Lecture 6: Finding Features (part 1/2) Lecture 6: Finding Features (part 1/2) Professor Fei- Fei Li Stanford Vision Lab Lecture 6 -! 1 What we will learn today? Local invariant features MoHvaHon Requirements, invariances Keypoint localizahon

More information

Edge Detection. Computer Vision P. Schrater Spring 2003

Edge Detection. Computer Vision P. Schrater Spring 2003 Edge Detection Computer Vision P. Schrater Spring 2003 Simplest Model: (Canny) Edge(x) = a U(x) + n(x) U(x)? x=0 Convolve image with U and find points with high magnitude. Choose value by comparing with

More information

Orientation Map Based Palmprint Recognition

Orientation Map Based Palmprint Recognition Orientation Map Based Palmprint Recognition (BM) 45 Orientation Map Based Palmprint Recognition B. H. Shekar, N. Harivinod bhshekar@gmail.com, harivinodn@gmail.com India, Mangalore University, Department

More information

SURVEY OF APPEARANCE-BASED METHODS FOR OBJECT RECOGNITION

SURVEY OF APPEARANCE-BASED METHODS FOR OBJECT RECOGNITION SURVEY OF APPEARANCE-BASED METHODS FOR OBJECT RECOGNITION Peter M. Roth and Martin Winter Inst. for Computer Graphics and Vision Graz University of Technology, Austria Technical Report ICG TR 01/08 Graz,

More information

Image matching. by Diva Sian. by swashford

Image matching. by Diva Sian. by swashford Image matching by Diva Sian by swashford Harder case by Diva Sian by scgbt Invariant local features Find features that are invariant to transformations geometric invariance: translation, rotation, scale

More information

SCALE-SPACE - Theory and Applications

SCALE-SPACE - Theory and Applications SCALE-SPACE - Theory and Applications Scale is embedded in the task: do you want only coins or TREASURE? SCALE-SPACE Theory and Applications - Scale-space theory is a framework of multiscale image/signal

More information

Given a feature in I 1, how to find the best match in I 2?

Given a feature in I 1, how to find the best match in I 2? Feature Matching 1 Feature matching Given a feature in I 1, how to find the best match in I 2? 1. Define distance function that compares two descriptors 2. Test all the features in I 2, find the one with

More information

arxiv: v1 [cs.cv] 10 Feb 2016

arxiv: v1 [cs.cv] 10 Feb 2016 GABOR WAVELETS IN IMAGE PROCESSING David Bařina Doctoral Degree Programme (2), FIT BUT E-mail: xbarin2@stud.fit.vutbr.cz Supervised by: Pavel Zemčík E-mail: zemcik@fit.vutbr.cz arxiv:162.338v1 [cs.cv]

More information

Machine vision. Summary # 4. The mask for Laplacian is given

Machine vision. Summary # 4. The mask for Laplacian is given 1 Machine vision Summary # 4 The mask for Laplacian is given L = 0 1 0 1 4 1 (6) 0 1 0 Another Laplacian mask that gives more importance to the center element is L = 1 1 1 1 8 1 (7) 1 1 1 Note that the

More information

6.869 Advances in Computer Vision. Prof. Bill Freeman March 1, 2005

6.869 Advances in Computer Vision. Prof. Bill Freeman March 1, 2005 6.869 Advances in Computer Vision Prof. Bill Freeman March 1 2005 1 2 Local Features Matching points across images important for: object identification instance recognition object class recognition pose

More information

ECE 468: Digital Image Processing. Lecture 8

ECE 468: Digital Image Processing. Lecture 8 ECE 68: Digital Image Processing Lecture 8 Prof. Sinisa Todorovic sinisa@eecs.oregonstate.edu 1 Point Descriptors Point Descriptors Describe image properties in the neighborhood of a keypoint Descriptors

More information

Machine vision, spring 2018 Summary 4

Machine vision, spring 2018 Summary 4 Machine vision Summary # 4 The mask for Laplacian is given L = 4 (6) Another Laplacian mask that gives more importance to the center element is given by L = 8 (7) Note that the sum of the elements in the

More information

Image Filtering. Slides, adapted from. Steve Seitz and Rick Szeliski, U.Washington

Image Filtering. Slides, adapted from. Steve Seitz and Rick Szeliski, U.Washington Image Filtering Slides, adapted from Steve Seitz and Rick Szeliski, U.Washington The power of blur All is Vanity by Charles Allen Gillbert (1873-1929) Harmon LD & JuleszB (1973) The recognition of faces.

More information

ITK Filters. Thresholding Edge Detection Gradients Second Order Derivatives Neighborhood Filters Smoothing Filters Distance Map Image Transforms

ITK Filters. Thresholding Edge Detection Gradients Second Order Derivatives Neighborhood Filters Smoothing Filters Distance Map Image Transforms ITK Filters Thresholding Edge Detection Gradients Second Order Derivatives Neighborhood Filters Smoothing Filters Distance Map Image Transforms ITCS 6010:Biomedical Imaging and Visualization 1 ITK Filters:

More information

Roadmap. Introduction to image analysis (computer vision) Theory of edge detection. Applications

Roadmap. Introduction to image analysis (computer vision) Theory of edge detection. Applications Edge Detection Roadmap Introduction to image analysis (computer vision) Its connection with psychology and neuroscience Why is image analysis difficult? Theory of edge detection Gradient operator Advanced

More information

Is the Scale Invariant Feature Transform (SIFT) really Scale Invariant?

Is the Scale Invariant Feature Transform (SIFT) really Scale Invariant? Is the Scale Invariant Feature Transform (SIFT) really Scale Invariant? Course prepared by: Jean-Michel Morel, Guoshen Yu and Ives Rey Otero October 24, 2010 Abstract This note is devoted to a mathematical

More information

Instance-level recognition: Local invariant features. Cordelia Schmid INRIA, Grenoble

Instance-level recognition: Local invariant features. Cordelia Schmid INRIA, Grenoble nstance-level recognition: ocal invariant features Cordelia Schmid NRA Grenoble Overview ntroduction to local features Harris interest points + SSD ZNCC SFT Scale & affine invariant interest point detectors

More information

TRACKING and DETECTION in COMPUTER VISION Filtering and edge detection

TRACKING and DETECTION in COMPUTER VISION Filtering and edge detection Technischen Universität München Winter Semester 0/0 TRACKING and DETECTION in COMPUTER VISION Filtering and edge detection Slobodan Ilić Overview Image formation Convolution Non-liner filtering: Median

More information

Motion Estimation (I) Ce Liu Microsoft Research New England

Motion Estimation (I) Ce Liu Microsoft Research New England Motion Estimation (I) Ce Liu celiu@microsoft.com Microsoft Research New England We live in a moving world Perceiving, understanding and predicting motion is an important part of our daily lives Motion

More information

Object Recognition Using Local Characterisation and Zernike Moments

Object Recognition Using Local Characterisation and Zernike Moments Object Recognition Using Local Characterisation and Zernike Moments A. Choksuriwong, H. Laurent, C. Rosenberger, and C. Maaoui Laboratoire Vision et Robotique - UPRES EA 2078, ENSI de Bourges - Université

More information

Instance-level l recognition. Cordelia Schmid INRIA

Instance-level l recognition. Cordelia Schmid INRIA nstance-level l recognition Cordelia Schmid NRA nstance-level recognition Particular objects and scenes large databases Application Search photos on the web for particular places Find these landmars...in

More information

KAZE Features. 1 Introduction. Pablo Fernández Alcantarilla 1, Adrien Bartoli 1, and Andrew J. Davison 2

KAZE Features. 1 Introduction. Pablo Fernández Alcantarilla 1, Adrien Bartoli 1, and Andrew J. Davison 2 KAZE Features Pablo Fernández Alcantarilla 1, Adrien Bartoli 1, and Andrew J. Davison 2 1 ISIT-UMR 6284 CNRS, Université d Auvergne, Clermont Ferrand, France {pablo.alcantarilla,adrien.bartoli}@gmail.com

More information

Advanced Edge Detection 1

Advanced Edge Detection 1 Advanced Edge Detection 1 Lecture 4 See Sections 2.4 and 1.2.5 in Reinhard Klette: Concise Computer Vision Springer-Verlag, London, 2014 1 See last slide for copyright information. 1 / 27 Agenda 1 LoG

More information

Linear Diffusion and Image Processing. Outline

Linear Diffusion and Image Processing. Outline Outline Linear Diffusion and Image Processing Fourier Transform Convolution Image Restoration: Linear Filtering Diffusion Processes for Noise Filtering linear scale space theory Gauss-Laplace pyramid for

More information

Dimensionality Reduc1on

Dimensionality Reduc1on Dimensionality Reduc1on contd Aarti Singh Machine Learning 10-601 Nov 10, 2011 Slides Courtesy: Tom Mitchell, Eric Xing, Lawrence Saul 1 Principal Component Analysis (PCA) Principal Components are the

More information

Introduction to Computer Vision

Introduction to Computer Vision Introduction to Computer Vision Michael J. Black Sept 2009 Lecture 8: Pyramids and image derivatives Goals Images as functions Derivatives of images Edges and gradients Laplacian pyramids Code for lecture

More information

Face detection and recognition. Detection Recognition Sally

Face detection and recognition. Detection Recognition Sally Face detection and recognition Detection Recognition Sally Face detection & recognition Viola & Jones detector Available in open CV Face recognition Eigenfaces for face recognition Metric learning identification

More information

Instance-level recognition: Local invariant features. Cordelia Schmid INRIA, Grenoble

Instance-level recognition: Local invariant features. Cordelia Schmid INRIA, Grenoble nstance-level recognition: ocal invariant features Cordelia Schmid NRA Grenoble Overview ntroduction to local features Harris interest t points + SSD ZNCC SFT Scale & affine invariant interest point detectors

More information

Image Alignment and Mosaicing

Image Alignment and Mosaicing Image Alignment and Mosaicing Image Alignment Applications Local alignment: Tracking Stereo Global alignment: Camera jitter elimination Image enhancement Panoramic mosaicing Image Enhancement Original

More information

Feature Extraction and Image Processing

Feature Extraction and Image Processing Feature Extraction and Image Processing Second edition Mark S. Nixon Alberto S. Aguado :*авш JBK IIP AMSTERDAM BOSTON HEIDELBERG LONDON NEW YORK OXFORD PARIS SAN DIEGO SAN FRANCISCO SINGAPORE SYDNEY TOKYO

More information

Learning Eigenfunctions: Links with Spectral Clustering and Kernel PCA

Learning Eigenfunctions: Links with Spectral Clustering and Kernel PCA Learning Eigenfunctions: Links with Spectral Clustering and Kernel PCA Yoshua Bengio Pascal Vincent Jean-François Paiement University of Montreal April 2, Snowbird Learning 2003 Learning Modal Structures

More information

Interest Point Detection with Wavelet Maxima Lines

Interest Point Detection with Wavelet Maxima Lines Interest Point Detection with Wavelet Maxima Lines Christophe Damerval, Sylvain Meignen To cite this version: Christophe Damerval, Sylvain Meignen. Interest Point Detection with Wavelet Maxima Lines. 7.

More information

Lecture 04 Image Filtering

Lecture 04 Image Filtering Institute of Informatics Institute of Neuroinformatics Lecture 04 Image Filtering Davide Scaramuzza 1 Lab Exercise 2 - Today afternoon Room ETH HG E 1.1 from 13:15 to 15:00 Work description: your first

More information

Multimedia Databases. Previous Lecture. 4.1 Multiresolution Analysis. 4 Shape-based Features. 4.1 Multiresolution Analysis

Multimedia Databases. Previous Lecture. 4.1 Multiresolution Analysis. 4 Shape-based Features. 4.1 Multiresolution Analysis Previous Lecture Multimedia Databases Texture-Based Image Retrieval Low Level Features Tamura Measure, Random Field Model High-Level Features Fourier-Transform, Wavelets Wolf-Tilo Balke Silviu Homoceanu

More information

Interest Operators. All lectures are from posted research papers. Harris Corner Detector: the first and most basic interest operator

Interest Operators. All lectures are from posted research papers. Harris Corner Detector: the first and most basic interest operator Interest Operators All lectures are from posted research papers. Harris Corner Detector: the first and most basic interest operator SIFT interest point detector and region descriptor Kadir Entrop Detector

More information

Visual Object Recognition

Visual Object Recognition Visual Object Recognition Lecture 2: Image Formation Per-Erik Forssén, docent Computer Vision Laboratory Department of Electrical Engineering Linköping University Lecture 2: Image Formation Pin-hole, and

More information

Multimedia Databases. Wolf-Tilo Balke Philipp Wille Institut für Informationssysteme Technische Universität Braunschweig

Multimedia Databases. Wolf-Tilo Balke Philipp Wille Institut für Informationssysteme Technische Universität Braunschweig Multimedia Databases Wolf-Tilo Balke Philipp Wille Institut für Informationssysteme Technische Universität Braunschweig http://www.ifis.cs.tu-bs.de 4 Previous Lecture Texture-Based Image Retrieval Low

More information