(1) Feature-point matching by D.J.Duff for CompVis 2007.11.15 Online: Feature Point Matching Detection, Extraction.

Slides:



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

TP14 - Local features: detection and description Computer Vision, FCUP, 2014 Miguel Coimbra Slides by Prof. Kristen Grauman.
Low Complexity Keypoint Recognition and Pose Estimation Vincent Lepetit.
Templates, Image Pyramids, and Filter Banks Slides largely from Derek Hoeim, Univ. of Illinois.
Templates and Image Pyramids Computational Photography Derek Hoiem, University of Illinois 09/06/11.
Lecture 5 Template matching
Robust and large-scale alignment Image from
Patch Descriptors CSE P 576 Larry Zitnick
Lecture 6: Feature matching CS4670: Computer Vision Noah Snavely.
1 Interest Operators Find “interesting” pieces of the image –e.g. corners, salient regions –Focus attention of algorithms –Speed up computation Many possible.
(1) Feature-point matching by D.J.Duff for CompVis Online: Feature Point Matching Detection,
Object Recognition with Invariant Features n Definition: Identify objects or scenes and determine their pose and model parameters n Applications l Industrial.
Lecture 4: Feature matching
1 Interest Operator Lectures lecture topics –Interest points 1 (Linda) interest points, descriptors, Harris corners, correlation matching –Interest points.
Feature matching and tracking Class 5 Read Section 4.1 of course notes Read Shi and Tomasi’s paper on.
Computing motion between images
Feature extraction: Corners and blobs
Scale Invariant Feature Transform (SIFT)
Blob detection.
Lecture 3a: Feature detection and matching CS6670: Computer Vision Noah Snavely.
CS4670: Computer Vision Kavita Bala Lecture 7: Harris Corner Detection.
CS4670: Computer Vision Kavita Bala Lecture 8: Scale invariance.
Lecture 6: Feature matching and alignment CS4670: Computer Vision Noah Snavely.
Scale-Invariant Feature Transform (SIFT) Jinxiang Chai.
Overview Introduction to local features
Generic object detection with deformable part-based models
Computer vision.
05 - Feature Detection Overview Feature Detection –Intensity Extrema –Blob Detection –Corner Detection Feature Descriptors Feature Matching Conclusion.
Templates, Image Pyramids, and Filter Banks Computer Vision Derek Hoiem, University of Illinois 01/31/12.
1 Interest Operators Harris Corner Detector: the first and most basic interest operator Kadir Entropy Detector and its use in object recognition SIFT interest.
Overview Harris interest points Comparing interest points (SSD, ZNCC, SIFT) Scale & affine invariant interest points Evaluation and comparison of different.
Local invariant features Cordelia Schmid INRIA, Grenoble.
Lecture 4: Feature matching CS4670 / 5670: Computer Vision Noah Snavely.
Templates, Image Pyramids, and Filter Banks
Templates and Image Pyramids Computational Photography Derek Hoiem, University of Illinois 09/03/15.
CSCE 643 Computer Vision: Extractions of Image Features Jinxiang Chai.
Local invariant features 1 Thursday October 3 rd 2013 Neelima Chavali Virginia Tech.
Lecture 7: Features Part 2 CS4670/5670: Computer Vision Noah Snavely.
21 June 2009Robust Feature Matching in 2.3μs1 Simon Taylor Edward Rosten Tom Drummond University of Cambridge.
Local invariant features Cordelia Schmid INRIA, Grenoble.
COMP322/S2000/L171 Robot Vision System Major Phases in Robot Vision Systems: A. Data (image) acquisition –Illumination, i.e. lighting consideration –Lenses,
Templates, Image Pyramids, and Filter Banks
Autonomous Robots Vision © Manfred Huber 2014.
Edges.
Feature extraction: Corners and blobs. Why extract features? Motivation: panorama stitching We have two images – how do we combine them?
CSE 185 Introduction to Computer Vision Feature Matching.
Local features: detection and description
CS654: Digital Image Analysis
776 Computer Vision Jan-Michael Frahm Spring 2012.
Blob detection.
SIFT.
776 Computer Vision Jan-Michael Frahm Spring 2012.
Interest Points EE/CSE 576 Linda Shapiro.
Lecture 07 13/12/2011 Shai Avidan הבהרה: החומר המחייב הוא החומר הנלמד בכיתה ולא זה המופיע / לא מופיע במצגת.
Distinctive Image Features from Scale-Invariant Keypoints
Scale Invariant Feature Transform (SIFT)
Project 1: hybrid images
3D Vision Interest Points.
TP12 - Local features: detection and description
Scale and interest point descriptors
Feature description and matching
CSE 455 – Guest Lectures 3 lectures Contact Interest points 1
Local features and image matching
SIFT keypoint detection
SIFT.
CSE 185 Introduction to Computer Vision
Lecture VI: Corner and Blob Detection
Feature descriptors and matching
Presentation transcript:

(1) Feature-point matching by D.J.Duff for CompVis Online: Feature Point Matching Detection, Extraction & Matching Part 2 – Matching via template matching (Previous slides: Feature point detection Moravec detection Harris detection Scale space detection Feature point Extraction ).

(2) Feature-point matching by D.J.Duff for CompVis Online: Template matching – X-correlation ● Template matching by Cross-correlation  You have already learnt template matching with sum of square differences – SSD: There you were trying to minimize a dissimilarity score.  Here you will learn about template matching by normalized cross-correlation: minimizing a similarity score.

(3) Feature-point matching by D.J.Duff for CompVis Online: Template matching – X-correlation ● Cross-correlation:  In statistical parlance, the cross-correlation between two random variables is a measure of how closely the two variables vary together.  {A,B} = {1,2},{8,16},{3,6},{-5,-10} ● The magnitude of these variables vary together closely – they have a high positive cross-correlation.  {C,D} = {-1,2},{-8,16},{-3,6},{5,-10} ● The magnitude of these vary oppositely closely – they have a high negative cross-correlation.  {E,F} = {-1,2},{8,0},{-3,6},{-5,-10} ● These don’t vary closely – they have a low cross-correlation.

(4) Feature-point matching by D.J.Duff for CompVis Online: Template matching – X-correlation ● Cross-correlation: ● Normalized cross-correlation (-1<0<-1):  {A,B} = {1,2},{8,16},{3,6},{-5,-10} ● X(A,B) = 43 C(A,B) = 1.0  {C,D} = {-1,2},{-8,16},{-3,6},{5,-10} ● X(C,D) = -43 C(A,B) = -1.0  {E,F} = {-1,2},{8,0},{-3,6},{-5,-10} ● X(E,F) = 7.4 C(A,B) = 0.25

(5) Feature-point matching by D.J.Duff for CompVis Online: Template matching – X-correlation ● Lingo alert:  Cross-correlation is usually called Covariance  Normalized cross-correlation is usually called Correlation:

(6) Feature-point matching by D.J.Duff for CompVis Online: Template matching – X-correlation ● Take an image (I) ● Take a target area to be matched (T) ● For each point in the image (I)  Find the normalized cross-correlation:  Where: and ● Template matching with Cross-correlation

(7) Feature-point matching by D.J.Duff for CompVis Online: Template matching – X-correlation ● Pick a pixel to be matched (u,v) ● For each (x,y) in the template ● Multiply its value (minus the template average) with the equivalent value in the image (minus the image patch average). Add to running total. ● Then divide the total by the square root of the product of the variances of the image and template. ● How to calculate the cross-correlation function C = 0.86 here

(8) Feature-point matching by D.J.Duff for CompVis Online: Template matching – X-correlation ● Template matching with Cross-correlation  Q: It looks similar to convolution. How does this relate to convolution?  A: It is indeed very similar. What you learnt as convolution is actually almost the cross-correlation that you learnt here (you just don’t divide by N). But this that is yet another definition of cross-correlation, from signal detection theory…  …so how would you do cross-correlation in matlab? Matlab: help conv2 help corr2 The cross-correlation function at the point is a measure of the degree of match between template and that point in the image.

(9) Feature-point matching by D.J.Duff for CompVis Online: Template matching – X-correlation ● Template matching with Cross-correlation  But... ● How well would this template match the image? ● Answer: Undefined! Divide by zero! ● It’s the normalization factor’s fault! (Doesn’t affect non-normalized cross-correlation).

(10) Feature-point matching by D.J.Duff for CompVis Online: Template matching ● Problems with naïve template matching  And... ● How well would this template match the image? ● Answer: ● Not rotationally invariant.

(11) Feature-point matching by D.J.Duff for CompVis Online: Template matching ● Problems with naïve template matching  And... ● Problem: Rotational invariance. ● Possible solution: Match every possible rotation.

(12) Feature-point matching by D.J.Duff for CompVis Online: Template matching ● Problems with naïve template matching  And... ● Problem: Rotational invariance. ● Possible solution: Align orientations. Direction of fastest change Direction of slowest change  E.g. by gradient direction

(13) Feature-point matching by D.J.Duff for CompVis Online: Template matching ● Problems with naïve template matching  Also… ● How well would this template match the image? ● Answer: 0.6 ● Needs to handle scale changes better too. ● Could check multiple scales…

(14) Feature-point matching by D.J.Duff for CompVis Online: Template matching ● Naïve cross-correlation is not the only way to do template matching. ● More sophisticated methods have to be able to handle  Scale changes  Rotation  Affine transformation  Lighting changes  Etc ● More efficient approaches are needed. E.g.  Using pyramidal (down-sampled) images  Frequency domain filtering  Iterative refinement

(15) Feature-point matching by D.J.Duff for CompVis Online: Alternatives to template matching ● Template matching can be refined to be made very useful. ● But other approaches exist too. ● These approaches generalize template matching by extracting a description (or descriptor) of the target feature point so that the descriptor can be matched. For example:  Training a classifier from local image information  Extracting local image statistics ● E.g. histogram of spatial distribution of local intensity gradient (as per SIFT)  More?  This improvement can be made at the extraction phase.

(16) Feature-point matching by D.J.Duff for CompVis Online: Summary ● Criteria for a good feature point detector are:  Repeatably detectable under transforms: ● Scale ● Rotation ● Affine transformation ● Illumination ● Etc  Computationally efficient  Accurate at localisation  High detection rate  Robust to noise

(17) Feature-point matching by D.J.Duff for CompVis Online: Summary ● Criteria for a good feature point matcher are:  Consistent  Invariant to: ● Scale ● Rotation ● Illumination ● Affine transformation ● Etc  Robust to noise  Accurate