Feature matching,face detection, object recognition and camera Flashcards

1
Q

Matching strategy

A

-Strategy 1: Max distance
Match against all features withing a geometric distance.
Difficult to set thressholds.BruteForce

-Strategy 2: Nearest Neigbor
In feature space
Threshold again

-Strategy 3: Nearest Neighbor distance ration.
Nearest distance to second nearest

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

Matching performance

A

precision: tp/tp+fp

recall:tp/tp+fn

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

Face detection Challenges

A

-Huge nr of pixels
-Multiple locations and scales
-Faces unlikely event
-key elements (fast processing of nonface candidates and low false positive rate is required)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

Viola Jones 1. Haar Features

A

Fast feature evaluation based on integral images.

Adv: Fast + efficient at calculations
Dis:Memoriy + preprocessing

Uses Rectangle filters.

Local Features->substract sum of pixels in white areas from sum of pixels in black areas.
2,3,4 rectangle features -> huge number.

Coarse Features
Sensitive to edges,bars,other simple structures.
Efficient computationally->compesates coarseness

Huge number of combinations

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

Viola Jones 2. Weak learners on harr features. 3 BOOSTING

4 Cascading

A

2Works on number evaluated by Haar feature. Sets threshold on a single feature and separates positive and negative examples.

  1. Boosting. Build a strong classifier vombining several weak vlassifiers. Weighted sum of simple weak learners. weight~accuarcy of this classsifier

ADABOOST(init equal weights. Selects best thresholds foreach filter reweight)At each round best weak classifier found OMNK

4.Cascading
Weak learners divided in different stages applied in cascade
Each stage acts as a filter.
First stage discards a sample, prevents work.
False negative = failure
False positive = acceptale
Reweight classifiers after each stage. From simple to more xomplex on each stage,from hight to low positive rate.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

Types of segmentation

A

Semantic: class to each pixel
Instance: detect object instances separately

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

Template matching

A

Template is:
-designed to serve as a model
-formed after a model
-example instance

Template matching: Finds instances of templates in the image.
Similarity measure should be chosen.

Weak points:
1-Illumination changes: use edge maps instead and use ZNCC
2.Scale changes:match in several scale, HOugh, Scale Space
3.Rotation:
Match in rotated version
Hough

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

Types of template matching:

A

1.Rigid
a.Correlation:
Template placed in every possible position w/out rotation and scaling.
compares.
Sliding window

b. Generalized Hough Transform.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

Bag Of Words

A

Image and object classification
Decomposes complex patterns into (semi)
independent features

Words can be represented using features
– Exploit discriminative properties
– Exploit invariance properties
– Re-use an efficient description

  1. Extract features – keypoints and descriptors
  2. Clustering in the feature space (e.g., K-means)
  3. Codebook generation: each cluster generates
    a representative sample (e.g., centroid)

Image classification:
– Evaluate the occurrence of each word in the
codeword
– Classify based on histogram

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

Pinhole Camera Model

A

We get a perfect definition if only one ray per
point reaches the sensor.

Focal length
* Distance between optical center
and image plane

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

Camera Projection

A

[I|0] core of projection process

We need to map points projected onto the
image plane in the coordinates used for pixels From 𝑥, 𝑦 to 𝑢, 𝑣.

Metric distances are converted
to pixels using the pixel width 𝑤
and ℎ height

Conversion factors are usually
defined as
– 𝑘𝑢 =1/𝑤
– 𝑘𝑣 =1/ℎ

Mapping from 𝑥, 𝑦 to 𝑢, 𝑣
is obtained by translation and
scaling:
𝑢 = 𝑢0 + 𝑥𝑝/𝑤= 𝑢0 + 𝑘𝑢𝑥𝑝
𝑣 = 𝑣0 +𝑦𝑝/ℎ = 𝑣0 + 𝑘𝑣𝑦𝑝

Summarizing and applying a similar conversion for 𝑣
yields:
𝑢 = 𝑢0 + 𝑓𝑢 𝑋𝑝/𝑍𝑝
𝑣 = 𝑣0 + 𝑓𝑣 𝑌𝑝/𝑍𝑝.

Camera matrix: K[i|0]

5 Intrinsic parameters : ku kv u0 v0 f.

Rototranslation T = [R t 0 1] rotation + translation. Extrensic parameters. 3 parameters for translation and 3 for rotation.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

Camera and lenses

A
  1. Adding a lense
    sharp image vs light intensity.
    Dont need pinhole.

Thin Lens model - in pinhole.

Point at a given distance in focus, others circle of confusion.
Adding a barrier reduces Circle of confusion

Focal length:
1-Thin lense distance which parallel rays intersect.
2-Pinhole model: distance between pinhoe and sensor.

Field of View: Angle Percieved by camera.
2 * Alpha that is a point P is seen.
Depends on sensor size + focal length.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

Lense Distortion

A

Derivation from ideal behaviour.
1-Radial
Entity depends on distance of disorted point from image center.
Pincusshion and Barrel.
Radial can be analyzed analitycally. Polynomial approx.

2-Tangential
Nonideal alignment between lens and sensor.
Similar tp perspective

Chromatic Aberration
Dsipersion:refractive index depend on wavelength

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

Real Cameras

A

We lose:
Angels
Distance
Parallel Lines.

Role of lense: Gather more light. Need to be focused.

Large Field of View, Small Focal length, camera close.
Small Field of view, Large focal length, camera far.

CCD AND CMOS: Measure total energy.
Greyscale sensors.
Color sensed by
a)3chip color -> separate RGB color images
b)Single Chip-1 image with filters
c)Chip penetration

Bayer Pattern-interpolation to provide complete color info at each pixel

FOVEON-no need interpolate. no info lost

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

Real Camera incoming light.

A

Apreature
-fraction of focal length.
f/2 on 50m = aperrature 25 mm.
Small f means large apperature.
Small apperature = more depth of field.

Shutter speed.
Almos closed and much time = almost opened and less time.

Shorter exposure time: freeze motion

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

Camera calibration

A

Estimating camera parameters.
intrinsic or i+e.
Needed to measure projection characteristics of a camera.

Links 3d object and projection provided only intrinsic

WHY?
Remove lens distortion
estimate 3d structure from camera motion
estimate depth
measure planar objects
Distances.

How?
Take object of known shape+appearance
take pictures
analyze.

Calibration pattern - any object of known shape and dimensions. FInd in image a shape easy recognizable.
Collect N images, find corners. Init intrising parameters for K and disortion. Same for extrensic with nonlinear LSP.

To help homography:
from pbject to image plane.
4 point per view are free. MIN 2 views are needed.