1 Ecological Statistics and Perceptual Organization Charless Fowlkes work with David Martin and Jitendra Malik at University of California at Berkeley.

Slides:



Advertisements
Similar presentations
Recovering Human Body Configurations: Combining Segmentation and Recognition Greg Mori, Xiaofeng Ren, and Jitentendra Malik (UC Berkeley) Alexei A. Efros.
Advertisements

PERCEPTION Chapter 4.5. Gestalt Principles  Gestalt principles are based on the idea that the whole is greater than the sum of the parts.  These principles.
1 Computational Vision CSCI 363, Fall 2012 Lecture 35 Perceptual Organization II.
Qualifying Exam: Contour Grouping Vida Movahedi Supervisor: James Elder Supervisory Committee: Minas Spetsakis, Jeff Edmonds York University Summer 2009.
ADS lab NCKU1 Michael Maire, Pablo Arbelaez, Charless Fowlkes, and Jitendra Malik university of California, Berkeley – Berkeley university of California,
Gestalt Principles of Visual Perception
Computer Vision Group University of California Berkeley Ecological Statistics of Good Continuation: Multi-scale Markov Models for Contours Xiaofeng Ren.
1 Contours and Junctions in Natural Images Jitendra Malik University of California at Berkeley (with Jianbo Shi, Thomas Leung, Serge Belongie, Charless.
Ghunhui Gu, Joseph J. Lim, Pablo Arbeláez, Jitendra Malik University of California at Berkeley Berkeley, CA
MESA LAB Depth ordering Guimei Zhang MESA LAB MESA (Mechatronics, Embedded Systems and Automation) LAB School of Engineering, University of California,
Recognition using Regions CVPR Outline Introduction Overview of the Approach Experimental Results Conclusion.
Computer Vision Group University of California Berkeley 1 Learning Scale-Invariant Contour Completion Xiaofeng Ren, Charless Fowlkes and Jitendra Malik.
Learning to Detect Natural Image Boundaries Using Local Brightness, Color, and Texture Cues David R. Martin Charless C. Fowlkes Jitendra Malik.
Understanding Gestalt Cues and Ecological Statistics using a Database of Human Segmented Images Charless Fowlkes, David Martin and Jitendra Malik Department.
Problem Sets Problem Set 3 –Distributed Tuesday, 3/18. –Due Thursday, 4/3 Problem Set 4 –Distributed Tuesday, 4/1 –Due Tuesday, 4/15. Probably a total.
Segmentation Divide the image into segments. Each segment:
Abstract We present a model of curvilinear grouping using piecewise linear representations of contours and a conditional random field to capture continuity.
Measuring the Ecological Statistics of Figure-Ground Charless Fowlkes, David Martin, Jitendra Malik.
1 Learning to Detect Natural Image Boundaries David Martin, Charless Fowlkes, Jitendra Malik Computer Science Division University of California at Berkeley.
CVR05 University of California Berkeley 1 Familiar Configuration Enables Figure/Ground Assignment in Natural Scenes Xiaofeng Ren, Charless Fowlkes, Jitendra.
Berkeley Vision GroupNIPS Vancouver Learning to Detect Natural Image Boundaries Using Local Brightness,
CVR05 University of California Berkeley 1 Cue Integration in Figure/Ground Labeling Xiaofeng Ren, Charless Fowlkes, Jitendra Malik.
Measuring the Ecological Statistics of Figure-Ground Charless Fowlkes, David Martin, Jitendra Malik.
A Database of Human Segmented Natural Images and Two Applications David Martin, Charless Fowlkes, Doron Tal, Jitendra Malik UC Berkeley
Segmentation by Clustering Reading: Chapter 14 (skip 14.5) Data reduction - obtain a compact representation for interesting image data in terms of a set.
1 The Ecological Statistics of Grouping by Similarity Charless Fowlkes, David Martin, Jitendra Malik Computer Science Division University of California.
WORD-PREDICTION AS A TOOL TO EVALUATE LOW-LEVEL VISION PROCESSES Prasad Gabbur, Kobus Barnard University of Arizona.
MSRI University of California Berkeley 1 Recovering Human Body Configurations using Pairwise Constraints between Parts Xiaofeng Ren, Alex Berg, Jitendra.
On Measuring * the Ecological Validity of Local Figure-Ground Cues Charless Fowlkes, David Martin, Jitendra Malik Computer Science Division University.
1 Occlusions – the world is flat without them! : Learning-Based Methods in Vision A. Efros, CMU, Spring 2009.
1 How do ideas from perceptual organization relate to natural scenes?
Computer Vision Group University of California Berkeley 1 Cue Integration in Figure/Ground Labeling Xiaofeng Ren, Charless Fowlkes and Jitendra Malik.
Cue Integration in Figure/Ground Labeling Xiaofeng Ren, Charless Fowlkes and Jitendra Malik, U.C. Berkeley We present a model of edge and region grouping.
Heather Dunlop : Advanced Perception January 25, 2006
The Hilbert Problems of Computer Vision
Performance Evaluation of Grouping Algorithms Vida Movahedi Elder Lab - Centre for Vision Research York University Spring 2009.
The Three R’s of Vision Jitendra Malik.
By Andrea Rees. Gestalt Principles 1) Closure 2) Proximity 3) Similarity 4) Figure VISUAL PERCEPTION PRINCIPLES OVERVIEW Depth Principles Binocular 1)
Computer Vision Lecture 5. Clustering: Why and How.
1 Contours and Junctions in Natural Images Jitendra Malik University of California at Berkeley (with Jianbo Shi, Thomas Leung, Serge Belongie, Charless.
BY JESSIE PARKER VISUAL PERCEPTION PRINCIPLES. VISUAL PERCEPTION Visual perception is the ability to interpret the surrounding environment by processing.
Visual Grouping and Recognition Jitendra Malik University of California at Berkeley Jitendra Malik University of California at Berkeley.
Chapter 6 Section 2: Vision. What we See Stimulus is light –Visible light comes from sun, stars, light bulbs, & is reflected off objects –Travels in the.
Supervised Learning of Edges and Object Boundaries Piotr Dollár Zhuowen Tu Serge Belongie.
‘rules’ that we apply to visual information  to assist our organization and interpretation of the information in consistent meaningful ways.
PSEUDO-RELEVANCE FEEDBACK FOR MULTIMEDIA RETRIEVAL Seo Seok Jun.
Gestalt Principles of Design
Ilya Gurvich 1 An Undergraduate Project under the supervision of Dr. Tammy Avraham Conducted at the ISL Lab, CS, Technion.
Fundamentals of Sensation and Perception RECOGNIZING VISUAL OBJECTS ERIK CHEVRIER NOVEMBER 23, 2015.
A New Method for Crater Detection Heather Dunlop November 2, 2006.
1 Mathematic Morphology used to extract image components that are useful in the representation and description of region shape, such as boundaries extraction.
Machine Vision Edge Detection Techniques ENT 273 Lecture 6 Hema C.R.
VISUAL PERCEPTION PRINCIPLES By Mikayla. VISUAL PERCEPTION PRINCIPLES  Gestalt principles 1.Closure 2.Proximity 3.Similarity 4.Figure-ground  Depth.
Chapter 24: Perception April 20, Introduction Emphasis on vision Feature extraction approach Model-based approach –S stimulus –W world –f,
Computational Vision Jitendra Malik University of California, Berkeley.
Rich feature hierarchies for accurate object detection and semantic segmentation 2014 IEEE Conference on Computer Vision and Pattern Recognition Ross Girshick,
Crowd Detection and Analysis By David Zeng CE at CCNY Mentor: Professor Hao Tang Graduate Student Mentor: Greg Olmschenk.
9/30/ Cognitive Robotics1 Gestalt Perception Cognitive Robotics David S. Touretzky & Ethan Tira-Thompson Carnegie Mellon Spring 2006.
Edge Detection Images and slides from: James Hayes, Brown University, Computer Vision course Svetlana Lazebnik, University of North Carolina at Chapel.
Human Computer Interaction
Gestalt Perception Cognitive Robotics David S. Touretzky &
Miguel Tavares Coimbra
Segmentation by clustering: mean shift
GESTALT PRINCIPLES IN ART AND DESIGN
A Tutorial on HOG Human Detection
Perceiving and Recognizing Objects
Contours and Junctions in Natural Images
Saliency detection Donghun Yeo CV Lab..
Cognitive Processes PSY 334
Cognitive Processes PSY 334
Presentation transcript:

1 Ecological Statistics and Perceptual Organization Charless Fowlkes work with David Martin and Jitendra Malik at University of California at Berkeley

2 “ I stand at the window and see a house, trees, sky. Theoretically I might say there were 327 brightnesses and nuances of color. Do I have 327? No. I have sky, house, and trees.”

4 “ I stand at the window and see a house, trees, sky. Theoretically I might say there were 327 brightnesses and nuances of color. Do I have 327? No. I have sky, house, and trees.” Laws of Organization in Perceptual Forms Max Wertheimer (1923)

5 Perceptual Organization Grouping Figure/Ground

6

7 Grouping by proximity

8 Grouping by similarity

9 Grouping by similarity (of shape)

10 Size and Surroundedness

11 turnyourhead.com Familiarity / Meaningfulness

12 Convexity

13 Perceptual organization as a computational theory of vision?

14 How do these cues apply to real world images? How are different cues combined? Why does the visual system use these cues?

15 Ecological Validity Brunswik & Kamiya 1953: Gestalt rules reflect the structure of the natural world Attempted to validate the grouping rule of proximity of similars Brunswik was ahead of his time… we now have the tools. Egon Brunswik ( )

16 Strategy 1.Collect high-level ground-truth annotations for a large collection of images 2.Develop computational models of cues for perceptual organization calibrated to ground-truth training data 3.Measure cue statistics and evaluate the relative “power” of different cues

17

18 30 subjects, age ,458 person hours over 8 months 1,020 Corel images 11,595 Segmentations –color, gray, inverted/negated “You will be presented a photographic image. Divide the image into some number of segments, where the segments represent “things” or “parts of things” in the scene. The number of segments is up to you, as it depends on the image. Something between 2 and 30 is likely to be appropriate. It is important that all of the segments have approximately equal importance.”

19 Berkeley Segmentation DataSet [BSDS]

20 Scene Background Sky TreesShore Water Small Top LR Mermaid Foreground Rocks Base Land (a) (b) (c) Scene Background TreesShore Water Small Top LR Mermaid Foreground Rocks Base Land Scene Background TreesShore Water Small Top LR Mermaid Foreground Rocks Base Land Sky

21 Overview Grouping –Local Boundary Detection –Local Human Performance Figure/Ground –Local Figure/Ground Cues –Local Human Performance Discussion

22 Non-BoundariesBoundaries T BC

23 Gradient Features Brightness Gradient (BG) –Difference of brightness distributions Color Gradient (CG) –Difference of color distributions Texture Gradient (TG) –Difference of distributions of V1-like filter responses 1976 CIE L*a*b* color space Distributions are represented by smoothed histograms  r (x,y)

24 Local Boundary Detection Image Boundary Cues Model PbPb Brightness Color Texture Using training data to learn the posterior probability of a boundary P(b=1|x,y,  ) from local gradient information Logistic regression to combine cues Cue Combination Brightness Color Texture

25 Canny Pb HumanImage

26 Canny Pb Humans Image

27

28 Goal Fewer False Positives Fewer Misses

29 Recall = P(P b > t | H = 1) Precision P(H = 1 | P b > t)

30 How good are humans locally? Off-Boundary On-Boundary Algorithm: r = 9, Humans: r = {5,9,18} Fixation(2s) -> Patch(200ms) -> Mask(1s)

31 Man versus Machine:

32 Findings Texture gradient information is important for natural scenes Optimal local cue combination is achievable with a simple linear model Algorithm for performing local boundary detection which performs nearly as well as local humans (and better than traditional edge detectors).

33 Overview Grouping –Local Boundary Detection –Local Human Performance Figure/Ground –Local Figure/Ground Cues –Local Human Performance Discussion

34 Local Cues for Figure/Ground Assume we have a perfect segmentation Can we predict which region a contour belongs to based on its local shape? –Size –Convexity –Lower Region

35 Figure-Ground Labeling - start with 200 segmented images of natural scenes - boundaries labeled by at least 2 different human subjects - subjects agree on 88% of contours labeled

36 Size(p) = log(Area F / Area G ) Size and Surroundedness [Rubin 1921] G F p

37 Convexity(p) = log(Conv F / Conv G ) Conv G = percentage of straight lines that lie completely within region G p G F Convexity [Metzger 1953, Kanizsa and Gerbino 1976]

38 LowerRegion(p) = θ G Lower Region [Vecera, Vogel & Woodman 2002] θ center of mass

39 Size Lower Region Convexity

40 Figural regions tend to lie below ground regions

41 Figural regions tend to be convex

42 Figural regions tend to be small

43 “Upper Bounding” Local Performance Present human subjects with local shapes, seen through an aperture. ConfigurationConfiguration + Content

44

45

46

47 Findings Convexity, size and lower-region are ecologically valid. Boundary configuration is relatively weak compared to luminance content. Local judgments based on luminance content can be quite accurate.

48 How do these cues apply to real world images? How are different cues combined? Why does the visual system use these cues? Perceptual organization as a computational theory of vision

49 How do ideas from perceptual organization relate to natural scenes?

50 How do ideas from perceptual organization relate to natural scenes?

51 THE END