w7 gemini

Question 1

What is the main goal of stereo vision?

Answer 1

To infer depth from two or more images taken from different viewpoints.

Question 2

What are the two main sub-problems in stereo vision?

Answer 2

Correspondence and Reconstruction.

Question 3

What is the correspondence problem in stereo vision?

Answer 3

The challenge of finding matching points in the left and right images that correspond to the same 3D point in the scene.

Question 4

What are some of the stereo constraints that can be used to solve the correspondence problem?

Answer 4

Epipolar constraint
maximum disparity constraint
continuity constraint
uniqueness constraint
ordering constraint.

Question 5

Explain the epipolar constraint for coplanar cameras.

Answer 5

For coplanar cameras, corresponding points will lie on the same row (the ‘epipolar’ line) in both images.

Question 6

What is the maximum disparity constraint?

Answer 6

The length of the search region for a corresponding point is limited by the maximum expected disparity.

Question 7

Explain the continuity constraint in the context of stereo vision.

Answer 7

Neighbouring points in an image are likely to correspond to points at similar depths.

Question 8

What is the uniqueness constraint in stereo vision?

Answer 8

A location in one image should ideally match only a single location in the other image.

Question 9

What is the ordering constraint in stereo vision?

Answer 9

Matching points along corresponding epipolar lines should appear in the same order in both images.

Question 10

What are the basic requirements for solving the correspondence problem?

Answer 10

Most scene points must be visible in both images, and corresponding image regions should appear similar.

Question 11

Under what conditions do the basic requirements for correspondence tend to hold?

Answer 11

When the distance of the 3D point from the cameras is much larger than the baseline (z&raquo_space; B).

Question 12

What are the two main approaches to solving the stereo correspondence problem?

Answer 12

Correlation-based methods and Feature-based methods.

Question 13

What is the output of correlation-based methods for correspondence?

Answer 13

Dense disparity maps.

Question 14

What is the output of feature-based methods for correspondence?

Answer 14

Sparse disparity maps.

Question 15

What are the characteristics of a simple case of stereo vision with coplanar cameras?

Answer 15

Image planes of cameras are coplanar, focal lengths are equal, optical centers are at the same height, and optical axes are parallel.

Question 16

How is disparity calculated for coplanar cameras?

Answer 16

d = x’L - x’R

Question 17

What is the relationship between depth and disparity?

Answer 17

Depth is inversely proportional to disparity.

Question 18

What is a disparity map?

Answer 18

A field of disparity vectors for a pair of stereo images.

Question 19

What is the significance of the baseline in a stereo vision system?

Answer 19

It is the distance between the optical centers of the two cameras.

Question 20

What are the characteristics of a complex case of stereo vision with non-coplanar cameras?

Answer 20

Cameras are not aligned such that their image planes are coplanar.

Question 21

What is measured instead of distance for disparity in non-coplanar cameras?

Answer 21

Angles.

Question 22

What is the formula for disparity in non-coplanar cameras using angles?

Answer 22

Disparity = αL - αR

Question 23

What is the horopter?

Answer 23

A curve in 3D space where objects project to corresponding points on the retina with zero disparity.

Question 24

What are some other cues to depth besides stereo vision?

Answer 24

Binocular, Oculomotor, Monocular, and Motion cues.

Question 25

What are the oculomotor cues to depth?

Answer 25

Accommodation and Convergence.

Question 26

Explain accommodation as a depth cue.

Answer 26

The shape of the lens in the eye adjusts to focus on objects at different depths.

Question 27

Explain convergence as a depth cue.

Answer 27

The angle of rotation of the eyes needed to fixate on an object at a certain depth.

Question 28

What are some monocular cues to depth?

Answer 28

Interposition, Size familiarity, Texture gradients, Linear perspective, Aerial perspective, and Shading.

Question 29

Explain interposition as a depth cue.

Answer 29

When one object blocks the view of another, the occluding object is perceived as being closer.

Question 30

Explain size familiarity as a depth cue.

Answer 30

Knowing the typical size of an object helps in judging its distance.

Question 31

Explain texture gradients as a depth cue.

Answer 31

For uniformly textured surfaces, the texture elements appear smaller and more closely spaced with increasing distance.

Question 32

Explain linear perspective as a depth cue.

Answer 32

Parallel lines appear to converge as they recede into the distance.

Question 33

Explain aerial perspective as a depth cue.

Answer 33

Distant objects appear fuzzier and have lower luminance contrast and color saturation.

Question 34

Explain shading as a depth cue.

Answer 34

The distribution of light and shadow on objects provides information about their shape and depth.

Question 35

What are some motion-induced cues to depth?

Answer 35

Motion parallax, Optic Flow, Accretion and deletion, and Structure from motion.

Question 36

Explain motion parallax as a depth cue.

Answer 36

The apparent speed and direction of motion of objects differ based on their distance.

Question 37

Explain optic flow as a depth cue.

Answer 37

As an observer moves, the pattern of apparent motion of objects provides information about relative depth.

Question 38

Explain accretion and deletion as motion-induced depth cues.

Answer 38

As an observer moves, parts of a background object may become covered or uncovered by a foreground object.

Question 39

Explain structure from motion (kinetic depth) as a depth cue.

Answer 39

The perception of a 3D structure can be derived from the changing 2D projections of the object.

Question 40

What is epipolar geometry?

Answer 40

The geometry of the situation involving two cameras looking at the same scene.

Question 41

What is an epipole?

Answer 41

The projection of the optical center of one camera onto the image plane of the other camera.

Question 42

What are epipolar lines?

Answer 42

Lines in one image that correspond to a point in the other image.

Question 43

What are conjugated epipolar lines?

Answer 43

The epipolar lines generated by the same 3D point on the left and right image planes.

Question 44

What is the Epipolar Constraint in the context of general (non-coplanar) cameras?

Answer 44

Corresponding points must lie on conjugated epipolar lines.

Question 45

What is rectification in the context of stereo vision?

Answer 45

A transformation applied to the stereo image pair to make the epipolar lines parallel to the image rows.

Question 46

What knowledge is required for triangulation and rectification in stereo vision?

Answer 46

Knowledge of intrinsic and extrinsic camera parameters.

Question 47

Why is recovery of depth information important in computer vision?

Answer 47

For controlling movement, 3D reconstruction, and object recognition.

Question 48

What is the difference between coplanar and non-coplanar cameras in terms of epipolar lines after rectification?

Answer 48

After rectification, epipolar lines in both cases become horizontal.

Question 49

What is a limitation of stereo vision related to the field of view?

Answer 49

A stereo pair of cameras can only find depth for locations within the common field of view.

Question 50

How does the length of the baseline affect the common field of view and depth error in coplanar stereo vision?

Answer 50

A short baseline leads to a larger common field of view but a larger depth error.

Question 51

What is the convergence angle in non-coplanar stereo vision?

Answer 51

The angle between the optical axes of the two cameras when they are fixated on a point.

Question 52

How does the location of the horopter depend on the vergence angle?

Answer 52

The location of the horopter changes based on the convergence angle.

Question 53

What is the significance of retinal disparity in human vision?

Answer 53

Some cortical neurons are tuned to retinal disparity.

Question 54

What are 'crossed' and 'uncrossed' disparities in human vision?

Answer 54

Points nearer than the horopter create 'crossed' disparities, while points more distant create 'uncrossed' disparities.

Question 55

What are some limitations of the uniqueness constraint?

Answer 55

It doesn't hold when surfaces lie along a line-of-sight for one camera.

Question 56

What are some limitations of the ordering constraint?

Answer 56

It doesn't hold when objects have different depths, causing occlusions.

Question 57

Why is correspondence fundamentally ambiguous?

Answer 57

Because for a given point in one image, there can be multiple potential matching points in the other image.

Question 58

What are two reasons why some points in each image may have no corresponding points in the other image?

Answer 58

Due to occlusion or because the cameras might have different fields of view.