PASCAL Slides
-
Upload
david-paredes -
Category
Documents
-
view
223 -
download
0
Transcript of PASCAL Slides
-
8/13/2019 PASCAL Slides
1/19
Discriminatively Trained Mixtures
of Deformable Part Models
Pedro Felzenszwalb and Ross Girshick
University of Chicago
David McAllesterToyota Technological Institute at Chicago
Deva RamananUC Irvine
http://www.cs.uchicago.edu/~pff/latent
http://www.cs.uchicago.edu/~pff/latenthttp://www.cs.uchicago.edu/~pff/latenthttp://www.cs.uchicago.edu/~pff/latenthttp://www.cs.uchicago.edu/~pff/latenthttp://www.cs.uchicago.edu/~pff/latenthttp://www.cs.uchicago.edu/~pff/latent -
8/13/2019 PASCAL Slides
2/19
Model Overview
Mixture of deformable part models (pictorial structures) Each component has global template + deformable parts
Fully trained from bounding boxes alone
-
8/13/2019 PASCAL Slides
3/19
2 component bicycle model
root filterscoarse resolution
part filtersfiner resolution
deformationmodels
-
8/13/2019 PASCAL Slides
4/19
Object Hypothesis
Image pyramid HOG feature pyramid
Multiscale model captures features at two resolutions
Score of object hypothesisis sum of filter scores
minus deformation costs
Score of filter is dot
product of filter with HOG
features underneath it
-
8/13/2019 PASCAL Slides
5/19
Connection with linear classifier
concatenation of HOGfeatures and part
displacements and 0s
concatenation filters and
deformation parameters
root filter
part filter
def param
part filter
def param
...
root filter
part filter
def param
part filter
def param
...
w
}}w: model parametersz: latent variables: component label and
filter placements
score on detection windowxcan be written as
component 1
component 2
-
8/13/2019 PASCAL Slides
6/19
Latent SVM
Linear in w if z is fixed
Regularization Hinge loss
-
8/13/2019 PASCAL Slides
7/19
Latent SVM training
Non-convex optimization
Huge number of negative examples Convex if we fixzfor positiveexamples
Optimization:
- Initialize wand iterate:- Pick bestzfor each positive example- Optimize wvia gradient descent with data mining
-
8/13/2019 PASCAL Slides
8/19
Initializing w
For kcomponent mixture model:
Split examples into ksets based on bounding box aspect ratio
Learn kroot filters using standard SVM
- Training data: warped positive examples and randomwindows from negative images (Dalal & Triggs)
Initialize parts by selecting patches from root filters- Subwindows with strong coefficients- Interpolate to get higher resolution filters- Initialize spatial model using fixed spring constants
-
8/13/2019 PASCAL Slides
9/19
Car model
root filterscoarse resolution
part filtersfiner resolution
deformationmodels
-
8/13/2019 PASCAL Slides
10/19
Person model
root filters
coarse resolution
part filters
finer resolution
deformation
models
-
8/13/2019 PASCAL Slides
11/19
Bottle model
root filters
coarse resolution
part filters
finer resolution
deformation
models
-
8/13/2019 PASCAL Slides
12/19
Histogram of Gradient (HOG) features
Dalal & Triggs:
- Histogram gradient orientations in 8x8 pixel blocks (9 bins)- Normalize with respect to 4 different neighborhoods and truncate- 9 orientations * 4 normalizations = 36 features per block
PCA gives ~10 features that capture all information
- Fewer parameters, speeds up convolution, but costly projection at runtime Analytic projection: spans PCA subspace and easy to compute
- 9 orientations + 4 normalizations = 13 features
We also use 2*9 contrast sensitive features for 31 features total
-
8/13/2019 PASCAL Slides
13/19
Bounding box prediction
predict (x1, y1) and(x2, y2)from part locations
linear function trained using least-squares regression
(x1, y1) (x2, y2)
-
8/13/2019 PASCAL Slides
14/19
Context rescoring
Rescore a detection using context defined by all detections Let vibe the max score of detector for class iin the image
Let sbe the score of a particular detection
Let (x1,y1), (x2,y2)be normalized bounding box coordinates
f = (s, x1, y1, x2, y2, v1, v2... , v20)
Train class specific classifier
- f is positive example if true positive detection- f is negative example if false positive detection
-
8/13/2019 PASCAL Slides
15/19
Bicycle detection
-
8/13/2019 PASCAL Slides
16/19
More bicycles
False positives
-
8/13/2019 PASCAL Slides
17/19
Car
-
8/13/2019 PASCAL Slides
18/19
Person
Bottle Horse
-
8/13/2019 PASCAL Slides
19/19
Code
Source code for the system and modelstrained on PASCAL 2006, 2007 and 2008
data are available here:
http://www.cs.uchicago.edu/~pff/latent
http://www.cs.uchicago.edu/~pff/latenthttp://www.cs.uchicago.edu/~pff/latent