sliding windows – silver bullet or evolutionary deadend?

18
Perceptual and Sensory Augmented Computing Discussion Session: Sliding Windows Sliding Windows – Silver Bullet or Evolutionary Deadend? Alyosha Efros, Bastian Leibe, Krystian Mikolajczyk Sicily Workshop, Syracusa, 23.09.2006

Upload: powa

Post on 28-Jan-2016

45 views

Category:

Documents


0 download

DESCRIPTION

Sliding Windows – Silver Bullet or Evolutionary Deadend?. Alyosha Efros, Bastian Leibe, Krystian Mikolajczyk Sicily Workshop, Syracusa, 23.09.2006. What is a Sliding Window Approach?. Search over space and scale Detection as subwindow classification problem - PowerPoint PPT Presentation

TRANSCRIPT

Page 1: Sliding Windows –  Silver Bullet or Evolutionary Deadend?

Perc

ep

tual an

d S

en

sory

Au

gm

en

ted

Com

pu

tin

gD

iscu

ssio

n S

essio

n:

Slid

ing

Win

dow

s

Sliding Windows – Silver Bullet or Evolutionary Deadend?

Alyosha Efros, Bastian Leibe, Krystian Mikolajczyk

Sicily Workshop, Syracusa, 23.09.2006

Page 2: Sliding Windows –  Silver Bullet or Evolutionary Deadend?

2

Perc

ep

tual an

d S

en

sory

Au

gm

en

ted

Com

pu

tin

g

A. Efros, B. Leibe, K. Mikolajczyk

Dis

cu

ssio

n S

essio

n:

Slid

ing

Win

dow

sWhat is a Sliding Window Approach?

• Search over space and scale

• Detection as subwindow classification problem

• “In the absence of a more intelligent strategy, any global image classification approach can be converted into a localization approach by using a sliding-window search.”

... ...

Page 3: Sliding Windows –  Silver Bullet or Evolutionary Deadend?

3

Perc

ep

tual an

d S

en

sory

Au

gm

en

ted

Com

pu

tin

g

A. Efros, B. Leibe, K. Mikolajczyk

Dis

cu

ssio

n S

essio

n:

Slid

ing

Win

dow

sTask: Object Localization in Still Images

• What options do we have to choose from? Sliding window approaches

– Classification problem– [Papageorgiou&Poggio,’00], [Schneiderman&Kanade,’00],

[Viola&Jones,01], [Mikolajczyk et al.,’04], [Torralba et al.,’04], [Dalal&Triggs,’05], [Wu&Nevatia,’05], [Laptev,’06],…

Feature-transform based approaches– Part-based generative models, typically with a star

topology– [Fergus et al.,’03], [Leibe&Schiele,’04], [Fei-Fei et al.,’04],

[Felszenszwalb&Huttenlocher,’05], [Winn&Criminisi,’06], [Opelt et al.,’06], [Mikolajczyk et al.,’06],…

Massively parallel NN architectures– e.g. convolutional NNs– [LeCun et al.,’98], [Osadchy et al.,’04], [Garcia et al.,??],…

“Smart segmentation” based approaches– Localization based on robustified bottom-up segmentation– [Todorovic&Ahuja,’06], [Roth&Ommer,’06]

Page 4: Sliding Windows –  Silver Bullet or Evolutionary Deadend?

4

Perc

ep

tual an

d S

en

sory

Au

gm

en

ted

Com

pu

tin

g

A. Efros, B. Leibe, K. Mikolajczyk

Dis

cu

ssio

n S

essio

n:

Slid

ing

Win

dow

s

Sliding-Window Approaches

• Pros: Can draw from vast stock of ML methods.

Independence assumption between subwindows.– Makes classification easier.– Process can be parallelized.

Simple technique, can be tried out very easily.– No translation/scale invariance required in model.

There are methods to do it very fast.– Cascades with AdaBoost/SVMs

Good detection performance on many benchmark datasets.

– e.g. face detection, VOC challenges

Direct control over search range (e.g. on ground plane).

Page 5: Sliding Windows –  Silver Bullet or Evolutionary Deadend?

5

Perc

ep

tual an

d S

en

sory

Au

gm

en

ted

Com

pu

tin

g

A. Efros, B. Leibe, K. Mikolajczyk

Dis

cu

ssio

n S

essio

n:

Slid

ing

Win

dow

s

Sliding-Window Approaches

• Cons: Can draw from vast stock of ML methods…

…as long as they can be evaluated in a few ms.

Need to evaluate many subwindows (100’000s). Needs very fast & accurate classification Many training examples required,

often limited to low training resolution. Can only deal with relatively small occlusions.

Still need to fuse resulting detections Hard/suboptimal from binary classification output

Classification task often ill-defined– How to label half a car?

Difficult to deal with changing aspect ratios

Page 6: Sliding Windows –  Silver Bullet or Evolutionary Deadend?

6

Perc

ep

tual an

d S

en

sory

Au

gm

en

ted

Com

pu

tin

g

A. Efros, B. Leibe, K. Mikolajczyk

Dis

cu

ssio

n S

essio

n:

Slid

ing

Win

dow

sDuality to Feature-Based Approaches…

• How to find maxima in the Hough space efficiently?

• Maxima search = coarse-to-fine sliding window stage!

• Main differences: All features evaluated upfront (instead of in cascade). Generative model instead of discriminative classifier. Maxima search already performs detection fusion.

y

s

Binned accum. array

y

s

xRefinement

(MSME)

y

s

xCandidatemaxima

y

s

Hough votes

Page 7: Sliding Windows –  Silver Bullet or Evolutionary Deadend?

7

Perc

ep

tual an

d S

en

sory

Au

gm

en

ted

Com

pu

tin

g

A. Efros, B. Leibe, K. Mikolajczyk

Dis

cu

ssio

n S

essio

n:

Slid

ing

Win

dow

s

So What is Left to Oppose?

1. Feature-based vs. Window-based?

2. (Almost) exclusive use of discriminative methods

3. Low training resolutions

4. How to deal with changing aspect ratios?

Page 8: Sliding Windows –  Silver Bullet or Evolutionary Deadend?

8

Perc

ep

tual an

d S

en

sory

Au

gm

en

ted

Com

pu

tin

g

A. Efros, B. Leibe, K. Mikolajczyk

Dis

cu

ssio

n S

essio

n:

Slid

ing

Win

dow

s1. Feature-based vs. Window-based

• May be mainly an implementation trade-off Few, localized features feature-based evaluation

better Many, dense features window-based evaluation

better Noticed already by e.g. [Schneiderman,’04] The trade-offs may change as your method

develops…

y

s

Page 9: Sliding Windows –  Silver Bullet or Evolutionary Deadend?

9

Perc

ep

tual an

d S

en

sory

Au

gm

en

ted

Com

pu

tin

g

A. Efros, B. Leibe, K. Mikolajczyk

Dis

cu

ssio

n S

essio

n:

Slid

ing

Win

dow

s2. Exclusive Use of Discriminative Methods

BackprojectedHypotheses

Interest PointsMatched Codebook Entries

Probabilistic Voting

Segmentation

3D Voting Space(continuous)

x

y

s

Backprojection

of Maxima

p(figure)Probabilities

[Leibe & Schiele,04]

Gen. Modelinside!

Page 10: Sliding Windows –  Silver Bullet or Evolutionary Deadend?

10

Perc

ep

tual an

d S

en

sory

Au

gm

en

ted

Com

pu

tin

g

A. Efros, B. Leibe, K. Mikolajczyk

Dis

cu

ssio

n S

essio

n:

Slid

ing

Win

dow

sGenerative Models for Sliding Windows

• Continuous confidence scores Smoother maxima in hypothesis space Coarser sampling possible

Page 11: Sliding Windows –  Silver Bullet or Evolutionary Deadend?

11

Perc

ep

tual an

d S

en

sory

Au

gm

en

ted

Com

pu

tin

g

A. Efros, B. Leibe, K. Mikolajczyk

Dis

cu

ssio

n S

essio

n:

Slid

ing

Win

dow

sGenerative Models for Sliding Windows

• Continuous confidence scores Smoother maxima in hypothesis space Coarser sampling possible

• Backprojection capability Determine a hypothesis’s support in the image Resolve overlapping cases

Page 12: Sliding Windows –  Silver Bullet or Evolutionary Deadend?

12

Perc

ep

tual an

d S

en

sory

Au

gm

en

ted

Com

pu

tin

g

A. Efros, B. Leibe, K. Mikolajczyk

Dis

cu

ssio

n S

essio

n:

Slid

ing

Win

dow

sGenerative Models for Sliding Windows

• Continuous confidence scores Smoother maxima in hypothesis space Coarser sampling possible

• Backprojection capability Determine a hypothesis’s support in the image Resolve overlapping cases

• Easier to deal with partial occlusion Part-based models Reasoning about missing parts

Page 13: Sliding Windows –  Silver Bullet or Evolutionary Deadend?

13

Perc

ep

tual an

d S

en

sory

Au

gm

en

ted

Com

pu

tin

g

A. Efros, B. Leibe, K. Mikolajczyk

Dis

cu

ssio

n S

essio

n:

Slid

ing

Win

dow

sSliding Windows for Generative Models

• Apply cascade idea to generative models Discriminative training Evaluate most promising features first

Page 14: Sliding Windows –  Silver Bullet or Evolutionary Deadend?

14

Perc

ep

tual an

d S

en

sory

Au

gm

en

ted

Com

pu

tin

g

A. Efros, B. Leibe, K. Mikolajczyk

Dis

cu

ssio

n S

essio

n:

Slid

ing

Win

dow

sSliding Windows for Generative Models

• Apply cascade idea to generative models Discriminative training Evaluate most promising features first

• Direct control over search range Only need to evaluate positions in search corridor Only need to consider subset of features Easier to adapt to different geometry

(e.g. curved ground surface)

Should combine discriminative and generative elements!

x

s

y Search corridor

Page 15: Sliding Windows –  Silver Bullet or Evolutionary Deadend?

15

Perc

ep

tual an

d S

en

sory

Au

gm

en

ted

Com

pu

tin

g

A. Efros, B. Leibe, K. Mikolajczyk

Dis

cu

ssio

n S

essio

n:

Slid

ing

Win

dow

s

3. Low Training Resolutions

• Many current s-w detectors operate on tiny images Viola&Jones: 2424 pixels Torralba et al.: 3232 pixels Dalal&Triggs: 6496 pixels (notable exception)

• Main reasons Training efficiency (exhaustive feature selection in

AdaBoost) Evaluation speed Want to recognize objects at small scales

• But… Limited information content available at those resolutions Not enough support to compensate for occlusions!

Page 16: Sliding Windows –  Silver Bullet or Evolutionary Deadend?

16

Perc

ep

tual an

d S

en

sory

Au

gm

en

ted

Com

pu

tin

g

A. Efros, B. Leibe, K. Mikolajczyk

Dis

cu

ssio

n S

essio

n:

Slid

ing

Win

dow

s

4. Changing Aspect Ratios

• Sliding window requires fixed window size Basis for learning efficient cascade classifier

• How to deal with changing aspect ratios? Fixed window size Wastes training dimensions

Adapted window size Difficult to share features

“Squashed” views [Dalal&Triggs] Need to squash test image, too

Page 17: Sliding Windows –  Silver Bullet or Evolutionary Deadend?

A. Efros, B. Leibe, K. Mikolajczyk 17

• What is wrong with sliding window?Search complexity?

Page 18: Sliding Windows –  Silver Bullet or Evolutionary Deadend?

A. Efros, B. Leibe, K. Mikolajczyk 18

• Is there anything that cannot be done with sliding window?