Ilya Gurvich 1 An Undergraduate Project under the supervision of Dr. Tammy Avraham Conducted at the ISL Lab, CS, Technion.

Slides:



Advertisements
Similar presentations
Histograms of Oriented Gradients for Human Detection
Advertisements

Evaluating Color Descriptors for Object and Scene Recognition Koen E.A. van de Sande, Student Member, IEEE, Theo Gevers, Member, IEEE, and Cees G.M. Snoek,
November 12, 2013Computer Vision Lecture 12: Texture 1Signature Another popular method of representing shape is called the signature. In order to compute.
Human Identity Recognition in Aerial Images Omar Oreifej Ramin Mehran Mubarak Shah CVPR 2010, June Computer Vision Lab of UCF.
1 Challenge the future HON4D: Histogram of Oriented 4D Normals for Activity Recognition from Depth Sequences Omar Oreifej Zicheng Liu CVPR 2013.
Carolina Galleguillos, Brian McFee, Serge Belongie, Gert Lanckriet Computer Science and Engineering Department Electrical and Computer Engineering Department.
1 Texture Texture is a description of the spatial arrangement of color or intensities in an image or a selected region of an image. Structural approach:
Computer Vision Lecture 16: Texture
Chapter 8 Content-Based Image Retrieval. Query By Keyword: Some textual attributes (keywords) should be maintained for each image. The image can be indexed.
Computer Vision for Human-Computer InteractionResearch Group, Universität Karlsruhe (TH) cv:hci Dr. Edgar Seemann 1 Computer Vision: Histograms of Oriented.
Image Segmentation Image segmentation (segmentace obrazu) –division or separation of the image into segments (connected regions) of similar properties.
The Global Digital Elevation Model (GTOPO30) of Great Basin Location: latitude 38  15’ to 42  N, longitude 118  30’ to 115  30’ W Grid size: 925 m.
São Paulo Advanced School of Computing (SP-ASC’10). São Paulo, Brazil, July 12-17, 2010 Looking at People Using Partial Least Squares William Robson Schwartz.
WRSTA, 13 August, 2006 Rough Sets in Hybrid Intelligent Systems For Breast Cancer Detection By Aboul Ella Hassanien Cairo University, Faculty of Computer.
Learning to Detect Natural Image Boundaries Using Local Brightness, Color, and Texture Cues David R. Martin Charless C. Fowlkes Jitendra Malik.
Three-dimensional co-occurrence matrices & Gabor filters: Current progress Gray-level co-occurrence matrices Carl Philips Gabor filters Daniel Li Supervisor:
Non-Local Characterization of Scenery Images: Statistics, 3D Reasoning, and a Generative Model Tamar Avraham and Michael Lindenbaum Technion.
Beyond bags of features: Adding spatial information Many slides adapted from Fei-Fei Li, Rob Fergus, and Antonio Torralba.
Berkeley Vision GroupNIPS Vancouver Learning to Detect Natural Image Boundaries Using Local Brightness,
Image Categorization by Learning and Reasoning with Regions Yixin Chen, University of New Orleans James Z. Wang, The Pennsylvania State University Published.
Classification and application in Remote Sensing.
CS292 Computational Vision and Language Visual Features - Colour and Texture.
1 How do ideas from perceptual organization relate to natural scenes?
1 Ecological Statistics and Perceptual Organization Charless Fowlkes work with David Martin and Jitendra Malik at University of California at Berkeley.
Heather Dunlop : Advanced Perception January 25, 2006
Real-Time Face Detection and Tracking Using Multiple Cameras RIT Computer Engineering Senior Design Project John RuppertJustin HnatowJared Holsopple This.
Tal Mor  Create an automatic system that given an image of a room and a color, will color the room walls  Maintaining the original texture.
Entropy and some applications in image processing Neucimar J. Leite Institute of Computing
Computer vision.
Attribute Expression Using Gray Level Co-Occurrence Sipuikinene Angelo*, Marcilio Matos,Kurt J Marfurt ConocoPhillips School of Geology & Geophysics, University.
8D040 Basis beeldverwerking Feature Extraction Anna Vilanova i Bartrolí Biomedical Image Analysis Group bmia.bmt.tue.nl.
B. Krishna Mohan and Shamsuddin Ladha
Texture. Texture is an innate property of all surfaces (clouds, trees, bricks, hair etc…). It refers to visual patterns of homogeneity and does not result.
Texture We would like to thank Amnon Drory for this deck הבהרה : החומר המחייב הוא החומר הנלמד בכיתה ולא זה המופיע / לא מופיע במצגת.
Classifying Images with Visual/Textual Cues By Steven Kappes and Yan Cao.
Texture We would like to thank Amnon Drory for this deck הבהרה : החומר המחייב הוא החומר הנלמד בכיתה ולא זה המופיע / לא מופיע במצגת.
Automatic Minirhizotron Root Image Analysis Using Two-Dimensional Matched Filtering and Local Entropy Thresholding Presented by Guang Zeng.
Supervised Learning of Edges and Object Boundaries Piotr Dollár Zhuowen Tu Serge Belongie.
Automatic Image Annotation by Using Concept-Sensitive Salient Objects for Image Content Representation Jianping Fan, Yuli Gao, Hangzai Luo, Guangyou Xu.
Course 9 Texture. Definition: Texture is repeating patterns of local variations in image intensity, which is too fine to be distinguished. Texture evokes.
Gang WangDerek HoiemDavid Forsyth. INTRODUCTION APROACH (implement detail) EXPERIMENTS CONCLUSION.
Levels of Image Data Representation 4.2. Traditional Image Data Structures 4.3. Hierarchical Data Structures Chapter 4 – Data structures for.
Kylie Gorman WEEK 1-2 REVIEW. CONVERTING AN IMAGE FROM RGB TO HSV AND DISPLAY CHANNELS.
Image Emotional Semantic Query Based On Color Semantic Description Wei-Ning Wang, Ying-Lin Yu Department of Electronic and Information Engineering, South.
Image Classification for Automatic Annotation
Category Independent Region Proposals Ian Endres and Derek Hoiem University of Illinois at Urbana-Champaign.
A New Method for Crater Detection Heather Dunlop November 2, 2006.
Slides from Dr. Shahera Hossain
Feature Selction for SVMs J. Weston et al., NIPS 2000 오장민 (2000/01/04) Second reference : Mark A. Holl, Correlation-based Feature Selection for Machine.
Classifying Covert Photographs CVPR 2012 POSTER. Outline  Introduction  Combine Image Features and Attributes  Experiment  Conclusion.
1.Learn appearance based models for concepts 2.Compute posterior probabilities or Semantic Multinomial (SMN) under appearance models. -But, suffers from.
Machine Vision Edge Detection Techniques ENT 273 Lecture 6 Hema C.R.
More sliding window detection: Discriminative part-based models
Image Quality Measures Omar Javed, Sohaib Khan Dr. Mubarak Shah.
Recognition of biological cells – development
Texture.
Data Driven Attributes for Action Detection
Nonparametric Semantic Segmentation
Histogram—Representation of Color Feature in Image Processing Yang, Li
Project Implementation for ITCS4122
Cheng-Ming Huang, Wen-Hung Liao Department of Computer Science
DICOM 11/21/2018.
CS 1674: Intro to Computer Vision Scene Recognition
Computer Vision Lecture 16: Texture II
Local Binary Patterns (LBP)
Computer and Robot Vision I
Midterm Exam Closed book, notes, computer Similar to test 1 in format:
Grouping/Segmentation
Midterm Exam Closed book, notes, computer Similar to test 1 in format:
A Novel Smoke Detection Method Using Support Vector Machine
Presentation transcript:

Ilya Gurvich 1 An Undergraduate Project under the supervision of Dr. Tammy Avraham Conducted at the ISL Lab, CS, Technion

2

What is it? 3

sky trees hill brushes trees river water trees 4

5

 “Ourdoor” LabelMe category.  Additional filtering of “open country”, “mountain” and “coast” images.  A total of 1144 images (256x256 pixels each).  These images are divided to an equally sized “training set” and a “testing set”.  Handling synonyms 6

 Use features vectors to represent patches  Use the multi-class SVM algorithm to learn the classes which these patches belong to  Find the optimal parameters for the SVM algorithm  Classify whole regions  This project is a part of a larger study in which global context was used 7

 The feature vector must be as discriminative as possible  Our feature vector contains a concatenation of: ◦ HSV Histogram ◦ Edges Directions Histogram / Histogram of Oriented Gradients (HoG) ◦ Gray-Level Co-occurrence Matrix (GLCM)  Based on Vogel & Scheile IJCV

 Each color channel (i.e. Hue, Saturation, Value) is used to build its histogram  These histograms are then concatenated 9

 The image is first converted to gray-scale  The Canny algorithm is then used to detect edges  For each pixel on which an edge is detected the direction of the gradient is calculated  The directions are then quantified and distributed to the histogram bins  The histogram is then normalized 10

 Used as an improvement to the Edges Directions Histogram  A gray-scale image is used  The directions and magnitudes of the gradients are calculated for every pixel of the image  The directions are quantified. Every pixel adds the gradient magnitude to the histogram bin determined by the direction  More formally: ◦ The value of a bin for the directions in the range [α,α+Δα] is: ◦ Where I is the image, is the gradient at the pixel p. 11

 Grey-Level Co-occurrence Matrix texture measurements have been the workhorse of image texture since they were proposed by Haralick in the 1970s.  Works on gray-scale images  Everyday texture terms - rough, silky, bumpy - refer to touch.  A texture that is rough to touch has: ◦ A large difference between high and low points, and ◦ A space between highs and lows approximately the same size as a finger.  Silky would have ◦ Little difference between high and low points, and ◦ The differences would be spaced very close together relative to finger size. Adapted from By Mryka Hall-Beyerhttp:// 12

 The GLCM is a tabulation of how often different combinations of pixel brightness values occur in an image.  The input to the GLCM computation algorithm a gray-scale image and a displacement vector (D).  The size of the GLCM matrix is NxN where N is the number of quantified gray-levels.  GLCM(i,j) counts the number of times that a pixel with the value of i was in the image and within an offset D from that pixel was a pixel with the value of j.  More formally: ◦ Where: (GLCM) i,j is the value of the GLCM matrix entry at (i,j). I is the image. R – the rows of the image. C – the columns of the image. I a,b is the gray-scale value at the pixel (i,j) in the image. 13

 We compute 4 GLCMs with the following displacements: ◦ (1,0), (1,1), (0,1), (-1,1)  We then calculate the following statistical measurements on each of the GLCMs: ◦ Contrast, Energy, Entropy, Homogeneity, Inverse Difference Moment, Correlation.  The 6 measurements per GLCM are then concatenated, forming a vector of 24 elements. 14

15

 A multi-class SVM algorithm with an RBF kernel is used to classify patches  A grid-search was performed to find the optimal SVM parameters: C and γ  The grid-search was implemented to execute parallelly in MATLAB  On a 4-core 2.5 GHz machine the search ran for 2 days 16

17

truth\predictionfieldgrassgroundlandmountainplainplantsriverrockssandseaskytreessnow field grass ground land mountain plain plants river rocks sand sea sky trees snow General accuracy rate: 71.76% 18

truth\predictionfieldgrassgroundlandmountainplainplantsriverrockssandseaskytreessnow field grass ground land mountain plain plants river rocks sand sea sky trees snow General accuracy rate: % 19

 The accuracy rates are correlated with the sizes of the classes ◦ Unbalanced dataset ◦ Learning the prior  Members of smaller classes are often confused with the semantically most similar larger class  Labeling noise  Upper bound on the accuracy rate of local patches 20

 Using the HSI+GLCM+Edges Feature Vector  Every region contains several patches  Associating a region to a category/class gives us a more global knowledge about the scene  Two voting methods ◦ A single vote per patch ◦ A weighted vote per patch, according to its probability (an output of the probabilistic SVM)  Will this improve the accuracy rates? Remember that there are usually several patches that form a region. 21

truth\predictionfieldgrassgroundlandmountainplainplantsriverrockssandseaskytreessnow field grass ground land mountain plain plants river rocks sand sea sky trees snow General accuracy rate: 70.77% 22

truth\predictionfieldgrassgroundlandmountainplainplantsriverrockssandseaskytreessnow field grass ground land mountain plain plants river rocks sand sea sky trees snow General accuracy rate: 71.34% 23

 The project was combined with: ◦ Non-Local Characterization of Scenery Images: Statistics, 3D Reasoning, and a Generative Model / Tamar Avraham and Michael Lindenbaum  Submitted to CVPR 2011: ◦ Multiple Region Classification for Scenery Images using Top-Bottom Order and Boundary Shape Cues  The following are now taken into account: ◦ The relative location of the region ◦ The height of the region ◦ The boundary between the regions ◦ Texture and color 24

sky mountain sea? ground? rocks?plants? only layout sky? sea? mountain? ground? sea rocks only color&texture + = sky mountain sea rocks 25  Goal: to show that region classification using global + local descriptors is better than only local descriptors

26 top bottom skytreesgroundsea

Ground truth Input image Relative location Boundary shape Color & texture All cues 27

19 categories! 28  Accuracy per class: ◦ Color & texture: higher accuracy for trees, field, rocks, plants, snow ◦ Layout: better for sky, mountain, sea, sand ◦ Other classes performance: very low due to their number. CueAccuracy Color&Texture0.615 Relative Location0.503 Boundary Shape0.452 Relative Loc. + Boundary Shape0.573 Color&Texture + Relative Loc Color&Texture + Boundary Shape0.641 All (ORC)0.682

sky sea river lake mountain cliff plateau land field valley bank beach sand ground rocks plants trees grass snow SKY WATER LAND SAND GROUND ROCKS PLANTS TREES GRASS SNOW MOUNTAIN PLAIN VALLEY BANK land structure land cover basic classeshigh level categories

30 ground truthInput image M-ORC

31

 Scenery images  Feature vectors  Optimal parameters  Patches classification  Regions classification  Incorporating global context 32

 Segmentation  Scene categorization  Extension to other domains  Picture alignment 33

34

35