Object Recognition - Columbia Universityxlx/ee4830/notes/lec11.pdfpattern classifier from examples...
Transcript of Object Recognition - Columbia Universityxlx/ee4830/notes/lec11.pdfpattern classifier from examples...
![Page 1: Object Recognition - Columbia Universityxlx/ee4830/notes/lec11.pdfpattern classifier from examples 13 goal: given x, infer y learning from examples: supervised learning given (x i,](https://reader036.fdocuments.us/reader036/viewer/2022081621/61224c94bf4aa03d5959ea58/html5/thumbnails/1.jpg)
1
Object Recognition
Lecture 11, April 20th, 2009
Lexing Xie
EE4830 Digital Image Processing http://www.ee.columbia.edu/~xlx/ee4830/
![Page 2: Object Recognition - Columbia Universityxlx/ee4830/notes/lec11.pdfpattern classifier from examples 13 goal: given x, infer y learning from examples: supervised learning given (x i,](https://reader036.fdocuments.us/reader036/viewer/2022081621/61224c94bf4aa03d5959ea58/html5/thumbnails/2.jpg)
2
Announcements
� HW#5 due today
� HW#6� last HW of the semester
� Problems 1-3 (100 pts + bonus) Due May 5th
� Covers object recognition and image compression
� Two written problems
� one programming with code skeleton given (rewards best recognition accuracy in class)
� Problems 4-5 (bonus points only) due May 7th
� Covers image reconstruction (lecture 13)
� Two written problems
![Page 3: Object Recognition - Columbia Universityxlx/ee4830/notes/lec11.pdfpattern classifier from examples 13 goal: given x, infer y learning from examples: supervised learning given (x i,](https://reader036.fdocuments.us/reader036/viewer/2022081621/61224c94bf4aa03d5959ea58/html5/thumbnails/3.jpg)
3
Roadmap to Date
Spatial Domain processing and enhancement
Image Transform and Filtering
Morphological Processing
Image Descriptors
Image Segmentation
Applications:Image Restoration, Object recognition, Image
Compression, Indexing and Retrieval, Reconstruction
![Page 4: Object Recognition - Columbia Universityxlx/ee4830/notes/lec11.pdfpattern classifier from examples 13 goal: given x, infer y learning from examples: supervised learning given (x i,](https://reader036.fdocuments.us/reader036/viewer/2022081621/61224c94bf4aa03d5959ea58/html5/thumbnails/4.jpg)
4
Lecture Outline
Problem: object recognition from images.
� What and why
� Pattern recognition primer
� Object recognition in controlled environments
� State of the art object recognition systems
![Page 5: Object Recognition - Columbia Universityxlx/ee4830/notes/lec11.pdfpattern classifier from examples 13 goal: given x, infer y learning from examples: supervised learning given (x i,](https://reader036.fdocuments.us/reader036/viewer/2022081621/61224c94bf4aa03d5959ea58/html5/thumbnails/5.jpg)
5
What is Object Recognition?
perceptibleperceptible materialmaterial
thingthing
Courtesy of http://people.csail.mit.edu/torralba/iccv2005/
![Page 6: Object Recognition - Columbia Universityxlx/ee4830/notes/lec11.pdfpattern classifier from examples 13 goal: given x, infer y learning from examples: supervised learning given (x i,](https://reader036.fdocuments.us/reader036/viewer/2022081621/61224c94bf4aa03d5959ea58/html5/thumbnails/6.jpg)
6
What is Object Recognition?
Sensory data Descriptions
Apple
+
One of the fundamental problems of computer vision:
Color, texture, shape, motion, size, weight, smell, touch, sound, …
“toy”, “stuffed Pooh”, “a frontal, close-up shot of stuffed Pooh”, “ToysRus#345812”, …
![Page 7: Object Recognition - Columbia Universityxlx/ee4830/notes/lec11.pdfpattern classifier from examples 13 goal: given x, infer y learning from examples: supervised learning given (x i,](https://reader036.fdocuments.us/reader036/viewer/2022081621/61224c94bf4aa03d5959ea58/html5/thumbnails/7.jpg)
7
Why?
� Science
� How do we recognize objects?
� Practice
� Robot navigation
� Medical diagnosis
� Security
� Industrial inspection and automation
� Human-computer interface
� Information retrieval
� …
![Page 8: Object Recognition - Columbia Universityxlx/ee4830/notes/lec11.pdfpattern classifier from examples 13 goal: given x, infer y learning from examples: supervised learning given (x i,](https://reader036.fdocuments.us/reader036/viewer/2022081621/61224c94bf4aa03d5959ea58/html5/thumbnails/8.jpg)
8
Applications of Object Recognition
Some images from http://www.cs.utexas.edu/~grauman/research/research.html
![Page 9: Object Recognition - Columbia Universityxlx/ee4830/notes/lec11.pdfpattern classifier from examples 13 goal: given x, infer y learning from examples: supervised learning given (x i,](https://reader036.fdocuments.us/reader036/viewer/2022081621/61224c94bf4aa03d5959ea58/html5/thumbnails/9.jpg)
9
![Page 10: Object Recognition - Columbia Universityxlx/ee4830/notes/lec11.pdfpattern classifier from examples 13 goal: given x, infer y learning from examples: supervised learning given (x i,](https://reader036.fdocuments.us/reader036/viewer/2022081621/61224c94bf4aa03d5959ea58/html5/thumbnails/10.jpg)
10
Lecture Outline
� Object recognition: what and why
� Object recognition in controlled environments
� Distance-based classifiers
� generalized linear classifiers
� Neural networks
� Bayes classifiers
� Object recognition in practice
� General object recognition systems
� Summary
![Page 11: Object Recognition - Columbia Universityxlx/ee4830/notes/lec11.pdfpattern classifier from examples 13 goal: given x, infer y learning from examples: supervised learning given (x i,](https://reader036.fdocuments.us/reader036/viewer/2022081621/61224c94bf4aa03d5959ea58/html5/thumbnails/11.jpg)
11
Objects as Vectors …
![Page 12: Object Recognition - Columbia Universityxlx/ee4830/notes/lec11.pdfpattern classifier from examples 13 goal: given x, infer y learning from examples: supervised learning given (x i,](https://reader036.fdocuments.us/reader036/viewer/2022081621/61224c94bf4aa03d5959ea58/html5/thumbnails/12.jpg)
12image vector representation
![Page 13: Object Recognition - Columbia Universityxlx/ee4830/notes/lec11.pdfpattern classifier from examples 13 goal: given x, infer y learning from examples: supervised learning given (x i,](https://reader036.fdocuments.us/reader036/viewer/2022081621/61224c94bf4aa03d5959ea58/html5/thumbnails/13.jpg)
13
pattern classifier from examples
� goal: given x, infer y
� learning from examples: supervised learning� given (xi, yi=f(xi)), i=1,…,N for some unknown function f
� find a “good approximation” to f
� rules versus data� encode human knowledge as rules
� e.g. the petal length and width of iris
� appropriate scenarios for supervised learning� no human expert (predict strength to cure AIDS given new molecule structure)
� human can perform task but can’t describe how they do it (e.g. handwriting recognition, object recognition)
� the desired function is changing constantly w.r.t. time, or user (stock trading decisions, user-specific spam filtering)
![Page 14: Object Recognition - Columbia Universityxlx/ee4830/notes/lec11.pdfpattern classifier from examples 13 goal: given x, infer y learning from examples: supervised learning given (x i,](https://reader036.fdocuments.us/reader036/viewer/2022081621/61224c94bf4aa03d5959ea58/html5/thumbnails/14.jpg)
14
minimum distance classifier
step 1: calculate “class prototypes” as the means
step 2: use the prototypes to classify a new example
“discriminant” function f:
![Page 15: Object Recognition - Columbia Universityxlx/ee4830/notes/lec11.pdfpattern classifier from examples 13 goal: given x, infer y learning from examples: supervised learning given (x i,](https://reader036.fdocuments.us/reader036/viewer/2022081621/61224c94bf4aa03d5959ea58/html5/thumbnails/15.jpg)
15
nearest neighbor classifier
� steps:� store all training examples
� classify a new example x? by finding the training example (xi, yi) that’s nearest to x? according to Euclidean distance, and copying the labels
![Page 16: Object Recognition - Columbia Universityxlx/ee4830/notes/lec11.pdfpattern classifier from examples 13 goal: given x, infer y learning from examples: supervised learning given (x i,](https://reader036.fdocuments.us/reader036/viewer/2022081621/61224c94bf4aa03d5959ea58/html5/thumbnails/16.jpg)
16
nearest neighbor classifier
“discriminant” function f: gray area -1; white area +1
� (implicit) decision boundaries form a subset of the Voronoi diagram of the training data – each line segment is equidistant between two points
� comments� conditioned on the distance metric
� prone to noisy, poorly scaled features
� can “smooth” the decision by looking at K-neighbors and vote
� good news: kNN is “universally consistent”
![Page 17: Object Recognition - Columbia Universityxlx/ee4830/notes/lec11.pdfpattern classifier from examples 13 goal: given x, infer y learning from examples: supervised learning given (x i,](https://reader036.fdocuments.us/reader036/viewer/2022081621/61224c94bf4aa03d5959ea58/html5/thumbnails/17.jpg)
17
linear classifier
� two desirables
� explicit (linear) decision boundary
� use many training examples/prototypes but do not need to remember all
![Page 18: Object Recognition - Columbia Universityxlx/ee4830/notes/lec11.pdfpattern classifier from examples 13 goal: given x, infer y learning from examples: supervised learning given (x i,](https://reader036.fdocuments.us/reader036/viewer/2022081621/61224c94bf4aa03d5959ea58/html5/thumbnails/18.jpg)
18
the perceptron algorithm
� learning a linear classifier
� given training data (xi, yi) and loss function
� find: weight vector [w; b] that minimizes expected loss on training data
� start from initial weights w0
� compute gradient
� update
� repeat until convergence
η : learning rate
use hinge loss:
![Page 19: Object Recognition - Columbia Universityxlx/ee4830/notes/lec11.pdfpattern classifier from examples 13 goal: given x, infer y learning from examples: supervised learning given (x i,](https://reader036.fdocuments.us/reader036/viewer/2022081621/61224c94bf4aa03d5959ea58/html5/thumbnails/19.jpg)
19
computing the gradient
compute gradient given
� η must decrease to zero in order to guarantee convergence.
� some algorithms (Newton’s) can automatically select η.
� local minimum is the global minimum for hinge loss
contribution from each training sample
contribution from each dimension of each training sample
![Page 20: Object Recognition - Columbia Universityxlx/ee4830/notes/lec11.pdfpattern classifier from examples 13 goal: given x, infer y learning from examples: supervised learning given (x i,](https://reader036.fdocuments.us/reader036/viewer/2022081621/61224c94bf4aa03d5959ea58/html5/thumbnails/20.jpg)
20
are all linear classifiers created equal?
� all of the separating hyper-planes have zero (hinge) loss
� the perceptron algorithm will stop as soon as
� may some hyper-planes more preferable to others
![Page 21: Object Recognition - Columbia Universityxlx/ee4830/notes/lec11.pdfpattern classifier from examples 13 goal: given x, infer y learning from examples: supervised learning given (x i,](https://reader036.fdocuments.us/reader036/viewer/2022081621/61224c94bf4aa03d5959ea58/html5/thumbnails/21.jpg)
21
Support Vector Machines
� Two key ideas:
� The “best” separating hyperplane has the largest margin.
� Class boundary can be linear in a higher-dimensional space, e.g.,
generalized linear discriminant
weighted (generalized) inner product with “support vectors”
![Page 22: Object Recognition - Columbia Universityxlx/ee4830/notes/lec11.pdfpattern classifier from examples 13 goal: given x, infer y learning from examples: supervised learning given (x i,](https://reader036.fdocuments.us/reader036/viewer/2022081621/61224c94bf4aa03d5959ea58/html5/thumbnails/22.jpg)
22
Neural Networks
![Page 23: Object Recognition - Columbia Universityxlx/ee4830/notes/lec11.pdfpattern classifier from examples 13 goal: given x, infer y learning from examples: supervised learning given (x i,](https://reader036.fdocuments.us/reader036/viewer/2022081621/61224c94bf4aa03d5959ea58/html5/thumbnails/23.jpg)
23
Neural Network Decision Boundaries
a single hidden layer, feed forward neural network is capable of approximating any continuous, multivariate function to any desired degree of accuracy and that failure to map a function arises from poor choice of network parameters, or an insufficient number of hidden neurons.
[Cybenko 1989]
![Page 24: Object Recognition - Columbia Universityxlx/ee4830/notes/lec11.pdfpattern classifier from examples 13 goal: given x, infer y learning from examples: supervised learning given (x i,](https://reader036.fdocuments.us/reader036/viewer/2022081621/61224c94bf4aa03d5959ea58/html5/thumbnails/24.jpg)
24
Digit Recognition with Neural Net
� LeCun et al, 1992, 1998, …http://yann.lecun.com/exdb/mnist/
0
1
![Page 25: Object Recognition - Columbia Universityxlx/ee4830/notes/lec11.pdfpattern classifier from examples 13 goal: given x, infer y learning from examples: supervised learning given (x i,](https://reader036.fdocuments.us/reader036/viewer/2022081621/61224c94bf4aa03d5959ea58/html5/thumbnails/25.jpg)
25
probabilistic classifiers
� what about probabilities
� p(x|y) is usually easy to obtain from training data
� can we estimate p(y|x) ?
![Page 26: Object Recognition - Columbia Universityxlx/ee4830/notes/lec11.pdfpattern classifier from examples 13 goal: given x, infer y learning from examples: supervised learning given (x i,](https://reader036.fdocuments.us/reader036/viewer/2022081621/61224c94bf4aa03d5959ea58/html5/thumbnails/26.jpg)
26
Bayes classifier
![Page 27: Object Recognition - Columbia Universityxlx/ee4830/notes/lec11.pdfpattern classifier from examples 13 goal: given x, infer y learning from examples: supervised learning given (x i,](https://reader036.fdocuments.us/reader036/viewer/2022081621/61224c94bf4aa03d5959ea58/html5/thumbnails/27.jpg)
27
Bayes classifier for Gaussian classes
![Page 28: Object Recognition - Columbia Universityxlx/ee4830/notes/lec11.pdfpattern classifier from examples 13 goal: given x, infer y learning from examples: supervised learning given (x i,](https://reader036.fdocuments.us/reader036/viewer/2022081621/61224c94bf4aa03d5959ea58/html5/thumbnails/28.jpg)
28
estimating the conditionals
� how do we estimate p(x|y)
� x1, x2, …, xN discrete:count over observed samplesto get the conditional histograms
�x1, x2, …, xN continuous and conditionally Gaussian
x scalar
![Page 29: Object Recognition - Columbia Universityxlx/ee4830/notes/lec11.pdfpattern classifier from examples 13 goal: given x, infer y learning from examples: supervised learning given (x i,](https://reader036.fdocuments.us/reader036/viewer/2022081621/61224c94bf4aa03d5959ea58/html5/thumbnails/29.jpg)
29
![Page 30: Object Recognition - Columbia Universityxlx/ee4830/notes/lec11.pdfpattern classifier from examples 13 goal: given x, infer y learning from examples: supervised learning given (x i,](https://reader036.fdocuments.us/reader036/viewer/2022081621/61224c94bf4aa03d5959ea58/html5/thumbnails/30.jpg)
30
Bayes classifier example
![Page 31: Object Recognition - Columbia Universityxlx/ee4830/notes/lec11.pdfpattern classifier from examples 13 goal: given x, infer y learning from examples: supervised learning given (x i,](https://reader036.fdocuments.us/reader036/viewer/2022081621/61224c94bf4aa03d5959ea58/html5/thumbnails/31.jpg)
31
![Page 32: Object Recognition - Columbia Universityxlx/ee4830/notes/lec11.pdfpattern classifier from examples 13 goal: given x, infer y learning from examples: supervised learning given (x i,](https://reader036.fdocuments.us/reader036/viewer/2022081621/61224c94bf4aa03d5959ea58/html5/thumbnails/32.jpg)
32
classification results
![Page 33: Object Recognition - Columbia Universityxlx/ee4830/notes/lec11.pdfpattern classifier from examples 13 goal: given x, infer y learning from examples: supervised learning given (x i,](https://reader036.fdocuments.us/reader036/viewer/2022081621/61224c94bf4aa03d5959ea58/html5/thumbnails/33.jpg)
33
exercise
P(x|ωj) being 1-D Gaussians with identical covariance.1) Towards which direction should the decision boundary move if p(ω1)>p(ω2)?Left/right/stay-put2) What if there is a third Gaussian?
![Page 34: Object Recognition - Columbia Universityxlx/ee4830/notes/lec11.pdfpattern classifier from examples 13 goal: given x, infer y learning from examples: supervised learning given (x i,](https://reader036.fdocuments.us/reader036/viewer/2022081621/61224c94bf4aa03d5959ea58/html5/thumbnails/34.jpg)
34
homework problem 2: classifying digits
� instruction/sample code available
� load digits from the MNIST dataset
� baseline 1-NN classifier
� experiment/observe/improve
� k-NN, with k=3, 5
� play with features space (PCA, …)
� Optionally experiment with other classifiers (SVM, neural net, …)
� compute error rate
� post examples that are correctly/incorrectly classified
� discuss what was tried and observed
![Page 35: Object Recognition - Columbia Universityxlx/ee4830/notes/lec11.pdfpattern classifier from examples 13 goal: given x, infer y learning from examples: supervised learning given (x i,](https://reader036.fdocuments.us/reader036/viewer/2022081621/61224c94bf4aa03d5959ea58/html5/thumbnails/35.jpg)
35
Lecture Outline
� object recognition: what and why
� object recognition as pattern classification
� general object recognition systems
� real-world challenges
� object recognition: a systems view
� current commercial systems
� survey of state-of the art
� demo websites
![Page 36: Object Recognition - Columbia Universityxlx/ee4830/notes/lec11.pdfpattern classifier from examples 13 goal: given x, infer y learning from examples: supervised learning given (x i,](https://reader036.fdocuments.us/reader036/viewer/2022081621/61224c94bf4aa03d5959ea58/html5/thumbnails/36.jpg)
36
Object Recognition End-to-End
Pre-processing
Feature Extraction
Sensor(s)
Classification
Post-processing
train classifier
Pre-processing
Feature Extraction
Sensor(s)
training testing
images
segments
features
classes
� window� parts or patches� salient points
� pixel vector� descriptors
� face? car?� statue of liberty?
� spatial/temporal smoothing� context, etc.
Recognition Target
![Page 37: Object Recognition - Columbia Universityxlx/ee4830/notes/lec11.pdfpattern classifier from examples 13 goal: given x, infer y learning from examples: supervised learning given (x i,](https://reader036.fdocuments.us/reader036/viewer/2022081621/61224c94bf4aa03d5959ea58/html5/thumbnails/37.jpg)
37
� Commercial object recognition
� Currently a $4 billion/year industry for inspection and assembly
� Almost entirely based on template matching
� Upcoming applications
� Mobile robots, toys, user interfaces
� Location recognition
� Digital camera panoramas, 3D scene modeling
Object Recognition in Practice
courtesy of David Lowe, website and CVPR 2003 Tutorial
![Page 38: Object Recognition - Columbia Universityxlx/ee4830/notes/lec11.pdfpattern classifier from examples 13 goal: given x, infer y learning from examples: supervised learning given (x i,](https://reader036.fdocuments.us/reader036/viewer/2022081621/61224c94bf4aa03d5959ea58/html5/thumbnails/38.jpg)
38
Industrial Applications
http://www.cs.ubc.ca/spider/lowe/vision.html
![Page 39: Object Recognition - Columbia Universityxlx/ee4830/notes/lec11.pdfpattern classifier from examples 13 goal: given x, infer y learning from examples: supervised learning given (x i,](https://reader036.fdocuments.us/reader036/viewer/2022081621/61224c94bf4aa03d5959ea58/html5/thumbnails/39.jpg)
39
http://www.appian-tech.com/
![Page 40: Object Recognition - Columbia Universityxlx/ee4830/notes/lec11.pdfpattern classifier from examples 13 goal: given x, infer y learning from examples: supervised learning given (x i,](https://reader036.fdocuments.us/reader036/viewer/2022081621/61224c94bf4aa03d5959ea58/html5/thumbnails/40.jpg)
40
http://www.sportvision.com/
![Page 41: Object Recognition - Columbia Universityxlx/ee4830/notes/lec11.pdfpattern classifier from examples 13 goal: given x, infer y learning from examples: supervised learning given (x i,](https://reader036.fdocuments.us/reader036/viewer/2022081621/61224c94bf4aa03d5959ea58/html5/thumbnails/41.jpg)
41
http://www.dipix.com/
![Page 42: Object Recognition - Columbia Universityxlx/ee4830/notes/lec11.pdfpattern classifier from examples 13 goal: given x, infer y learning from examples: supervised learning given (x i,](https://reader036.fdocuments.us/reader036/viewer/2022081621/61224c94bf4aa03d5959ea58/html5/thumbnails/42.jpg)
42
What to Recognize
Specific
Wild card Tower Bridge
Categories
building buildingbutterfly butterfly
The Stata Center
Kristen Grauman, http://www.cs.utexas.edu/~grauman/research/
![Page 43: Object Recognition - Columbia Universityxlx/ee4830/notes/lec11.pdfpattern classifier from examples 13 goal: given x, infer y learning from examples: supervised learning given (x i,](https://reader036.fdocuments.us/reader036/viewer/2022081621/61224c94bf4aa03d5959ea58/html5/thumbnails/43.jpg)
43
Recognize Specific Objects (1)
Appearance Matching
[Nayar, Murase et. al.]
![Page 44: Object Recognition - Columbia Universityxlx/ee4830/notes/lec11.pdfpattern classifier from examples 13 goal: given x, infer y learning from examples: supervised learning given (x i,](https://reader036.fdocuments.us/reader036/viewer/2022081621/61224c94bf4aa03d5959ea58/html5/thumbnails/44.jpg)
44
� PCA on the training set.
� Estimate parameters of a low-dimensional pose manifold with splines.
� Match new image to the closest point on the manifold.
![Page 45: Object Recognition - Columbia Universityxlx/ee4830/notes/lec11.pdfpattern classifier from examples 13 goal: given x, infer y learning from examples: supervised learning given (x i,](https://reader036.fdocuments.us/reader036/viewer/2022081621/61224c94bf4aa03d5959ea58/html5/thumbnails/45.jpg)
45
Recognize Specific Objects (2)
� Part-based approach
� Image content is transformed into local feature coordinates that are invariant to translation, rotation, scale, and other imaging parameters
� select “interest points” that are stable extrema points across different scales.
SIFT Features David Lowe, CVPR 2003 Tutorial
![Page 46: Object Recognition - Columbia Universityxlx/ee4830/notes/lec11.pdfpattern classifier from examples 13 goal: given x, infer y learning from examples: supervised learning given (x i,](https://reader036.fdocuments.us/reader036/viewer/2022081621/61224c94bf4aa03d5959ea58/html5/thumbnails/46.jpg)
46
SIFT Descriptor
� Thresholded image gradients are sampled over 16x16 array of locations in scale space (Gaussian-weighted).
� Create array of orientation histograms
� 8 orientations x 4x4 histogram array = 128 dimensions
David Lowe, CVPR 2003 Tutorial
![Page 47: Object Recognition - Columbia Universityxlx/ee4830/notes/lec11.pdfpattern classifier from examples 13 goal: given x, infer y learning from examples: supervised learning given (x i,](https://reader036.fdocuments.us/reader036/viewer/2022081621/61224c94bf4aa03d5959ea58/html5/thumbnails/47.jpg)
47
Object Category Recognition
![Page 48: Object Recognition - Columbia Universityxlx/ee4830/notes/lec11.pdfpattern classifier from examples 13 goal: given x, infer y learning from examples: supervised learning given (x i,](https://reader036.fdocuments.us/reader036/viewer/2022081621/61224c94bf4aa03d5959ea58/html5/thumbnails/48.jpg)
48
Overview of object category recognition …see iccv tutorial
![Page 49: Object Recognition - Columbia Universityxlx/ee4830/notes/lec11.pdfpattern classifier from examples 13 goal: given x, infer y learning from examples: supervised learning given (x i,](https://reader036.fdocuments.us/reader036/viewer/2022081621/61224c94bf4aa03d5959ea58/html5/thumbnails/49.jpg)
49
Demos
� Pittpatt http://demo.pittpatt.com/
![Page 50: Object Recognition - Columbia Universityxlx/ee4830/notes/lec11.pdfpattern classifier from examples 13 goal: given x, infer y learning from examples: supervised learning given (x i,](https://reader036.fdocuments.us/reader036/viewer/2022081621/61224c94bf4aa03d5959ea58/html5/thumbnails/50.jpg)
50It’s not just vision…
•8pm 10 Oct 05
•London
•3pm 10 Sep 05
•downloaded
from http://...
•10am 7 Sep 05
•Australian park
•Jim, Jill nearby
•4pm 8 Sep 05
•Sydney
“two koalas
seen on nat. park trip
with Jim and Jill”
“John and
his new car”
Integrate with mobile sensor information (GPS, time, nearby object or people), calendar, schedule…Infer semantically rich meta-data labels from joint sources.
“Jill and koala on
nat. park trip”
“office parking lot”
“car to consider
purchasing”
http://www.cs.utexas.edu/~grauman/research/research.html
![Page 51: Object Recognition - Columbia Universityxlx/ee4830/notes/lec11.pdfpattern classifier from examples 13 goal: given x, infer y learning from examples: supervised learning given (x i,](https://reader036.fdocuments.us/reader036/viewer/2022081621/61224c94bf4aa03d5959ea58/html5/thumbnails/51.jpg)
51
Summary
� The object recognition problem
� Pattern classification primer
� Object recognition grown up
� Readings: G&W 12.1-12.2
� Reference: Duda, Hart, Stork, “Pattern Classification”, 2nd Ed.
� Next time: Image Compression
Additional acknowledgements: Dan Ellis, EE6820 Slides; Duda, Hart& Stork, Pattern classificaion 2nd Ed., David Claus and Christoph F. Eick: Nearest Neighbor Editing and Condensing Techniques