Visual Category as Collection of filters
description
Transcript of Visual Category as Collection of filters
![Page 1: Visual Category as Collection of filters](https://reader036.fdocuments.us/reader036/viewer/2022070405/56813f05550346895da98880/html5/thumbnails/1.jpg)
Knowing a Good HOG Filter When You See It:
Efficient Selection of Filters for Detection
Ejaz Ahmed1, Gregory Shakhnarovich2, and Subhransu Maji3
1 University of Maryland, College Park
2 Toyota Technological Institute at Chicago
3 University of Massachusetts, Amherst
![Page 2: Visual Category as Collection of filters](https://reader036.fdocuments.us/reader036/viewer/2022070405/56813f05550346895da98880/html5/thumbnails/2.jpg)
Visual Category as Collection of fi lters
Poselets Mid Level Discriminative Patches
Exemplar SVMs
![Page 3: Visual Category as Collection of filters](https://reader036.fdocuments.us/reader036/viewer/2022070405/56813f05550346895da98880/html5/thumbnails/3.jpg)
Candidate Generation
Candidate Generation
Pool of Filters
Filters are generated using positives and negatives examples.
Generation of a large pool of fi lters.
Positives Negatives
Filter (SVM Classifier)
![Page 4: Visual Category as Collection of filters](https://reader036.fdocuments.us/reader036/viewer/2022070405/56813f05550346895da98880/html5/thumbnails/4.jpg)
Two sources of ineffi ciency
Candidate Selection
CandidateSelection
Selected Filters (n)
(n << N)
Pool of Filters (N)
Impractical to use all generated fi lters.
Redundancy
Noise
good
bad
![Page 5: Visual Category as Collection of filters](https://reader036.fdocuments.us/reader036/viewer/2022070405/56813f05550346895da98880/html5/thumbnails/5.jpg)
Candidate Selection Cont…
Expensive Evaluation
Selected Filters (n)
(n << N)
Run as detectorPool of Filters (N)
Bottleneck
![Page 6: Visual Category as Collection of filters](https://reader036.fdocuments.us/reader036/viewer/2022070405/56813f05550346895da98880/html5/thumbnails/6.jpg)
What we Propose
Expensive Evaluation
Selected Filters (n)
(n << N)
Run as detectorPool of Filters (N)
By passes explicit evaluation
fast
Our Contribution : fast automatic selection of a subset of discriminative and non redundant filters given a collection of filters
![Page 7: Visual Category as Collection of filters](https://reader036.fdocuments.us/reader036/viewer/2022070405/56813f05550346895da98880/html5/thumbnails/7.jpg)
Category Independent Model
Images +/-
Pool (Candidate Filters)N >> n
Selected Filters (n)
fast slow
(w , λ)
Test Category
fast
N candidates
n selected
fast
Can rank fi lters as accurately as a direct evaluation on thousands of examples.
![Page 8: Visual Category as Collection of filters](https://reader036.fdocuments.us/reader036/viewer/2022070405/56813f05550346895da98880/html5/thumbnails/8.jpg)
Poselets
Poselets are semantically aligned discriminative patterns that capture parts of object.
Patches are often far visually, but they are close semantically
![Page 9: Visual Category as Collection of filters](https://reader036.fdocuments.us/reader036/viewer/2022070405/56813f05550346895da98880/html5/thumbnails/9.jpg)
Poselet Architecture
Candidate Selection :
Candidate Generation :
24
76
Timings
generation
selection
Total Time = 20hrs
![Page 10: Visual Category as Collection of filters](https://reader036.fdocuments.us/reader036/viewer/2022070405/56813f05550346895da98880/html5/thumbnails/10.jpg)
Save significantly in training time if we can quickly select small set of relevant exemplars.
ESVMSVM for each positive example
Test time
Redundant Exemplars
![Page 11: Visual Category as Collection of filters](https://reader036.fdocuments.us/reader036/viewer/2022070405/56813f05550346895da98880/html5/thumbnails/11.jpg)
Good Filters Bad Filters
Good / Bad Filters
Gradient orientation within a cell (active simultaneously)
Gradient orientation of neighboring cells (lines, curves)
![Page 12: Visual Category as Collection of filters](https://reader036.fdocuments.us/reader036/viewer/2022070405/56813f05550346895da98880/html5/thumbnails/12.jpg)
Norm: consistent with high degree of alignment. Normalized Norm: Makes norm invariant to fi lter
dimension.
Cell Covariance: Diff erent orientation bins within a cell are highly structured. Gao et al. ECCV 2012
Cell Cross Covariance: Strong correlation between fi lter weights in nearby spatial locations.
Features for fi lter Ranking
Cell Covariance
Cell Cross Covariance
Decreasing Norm
![Page 13: Visual Category as Collection of filters](https://reader036.fdocuments.us/reader036/viewer/2022070405/56813f05550346895da98880/html5/thumbnails/13.jpg)
– representation of fi lter Goal : model ranking score of by a linear functionTraining data : ,
where is number of training categories. where N is number of filters per category. is estimated quality, obtained by expensive method.
is ordered in descending value of - , for measures how much better is from
Slack rescaled hinge loss
Learning to Rank Filters
![Page 14: Visual Category as Collection of filters](https://reader036.fdocuments.us/reader036/viewer/2022070405/56813f05550346895da98880/html5/thumbnails/14.jpg)
Selected parts should be individually good and complimentary.
First fi lter - fi lters selected so farSelect next fi lter using following
Greedy approximation for Diversity
Selected Filters
Not yet Selected
0.9
0.4
0.1
Added to selected set
![Page 15: Visual Category as Collection of filters](https://reader036.fdocuments.us/reader036/viewer/2022070405/56813f05550346895da98880/html5/thumbnails/15.jpg)
LDA Acceleration
(w , λ)
N SVM filters(Candidate Generation)
n Selected Filters(SVM)
N LDA filters(Candidate Generation)
(w , λ)
n Selected Filters(LDA)
n Selected Filters(SVM)
SVM bootstrappi
ng
Poor performan
ce
Goodperforman
ce
Goodperforman
ce
Selection with LDA Acceleration
SVM bootstrapping
LDA
Our Selection Method
![Page 16: Visual Category as Collection of filters](https://reader036.fdocuments.us/reader036/viewer/2022070405/56813f05550346895da98880/html5/thumbnails/16.jpg)
Experiments with Poselets
Test category
Filters used for training from remaining categories
800 poselet fi lters for each categoryGoal : given a category select 100 out of 800 fi ltersRanking taskDetection task
![Page 17: Visual Category as Collection of filters](https://reader036.fdocuments.us/reader036/viewer/2022070405/56813f05550346895da98880/html5/thumbnails/17.jpg)
Performance of Ranker
Predicted ranking vs true ranking as per AP scores.
Norm(svm)
Σ – Norm(svm)
Gao et al. ECCV 2012
Rank(lda)
Rank(svm)< < <
![Page 18: Visual Category as Collection of filters](https://reader036.fdocuments.us/reader036/viewer/2022070405/56813f05550346895da98880/html5/thumbnails/18.jpg)
1x
3x
8x
3x
3x
3x
8x
3x
3x
8x
2.4x
Detection Results
Σ – Norm (svm)
Random
Norm (svm)
10% Val
Norm (svm) + Div
Rank (svm)
Σ – Norm (svm) + Div
Oracle (expensive evaluation)
Rank (svm) + Div
Rank (lda) + Div
Rank (lda) + Div
2X Seeds
Increasin
g Perfo
rmance
Speed up w.r.t. OracleBy constructing a
poselet detector using selected fi lters
Order of magnitude Speed up.Improved performance than Oracle
![Page 19: Visual Category as Collection of filters](https://reader036.fdocuments.us/reader036/viewer/2022070405/56813f05550346895da98880/html5/thumbnails/19.jpg)
Each category has 630 exemplars on average.Goal select 100 exemplars such that they reproduce
result for optimal set of 100 exemplars.Optimal set – weights of each exemplar in the final
scoring model. (Oracle)Frequency of exemplars
Experiments with exemplar SVMs
Frequent Exemplar
Rare Exemplar
![Page 20: Visual Category as Collection of filters](https://reader036.fdocuments.us/reader036/viewer/2022070405/56813f05550346895da98880/html5/thumbnails/20.jpg)
We have presented an automatic mechanism for selecting diverse set of discriminative fi lters.
Order of magnitude improvement in training time.Our approach is applicable to any discriminative
architecture that uses a collection of fi lters. Insight into what makes a good fi lter for object
detection.Can be used as an attention mechanism during test
time Reduce number of convolutions / hashing lookups.
Bottom line: One can tell whether a fi lter is useful for a category without knowing what that category is, just by “looking” at the fi lter.