Scribd
Upload
34 views5 pages

CV Potential Questions

The document contains questions related to computer vision concepts like structure from motion, bundle adjustment, triangulation, fundamental matrix, and panoramic image processing. Bundle adjustment is now the standard method for structure from motion problems involving hundreds of weakly calibrated images and thousands of points. Bundle adjustment combined with gap closing can automatically assemble a panorama if the user takes overlapping images in sequence and specifies the start and end images. Professional panoramic photographers often use pan-tilt heads that make precise stops at specific rotation angles to aid in panorama stitching. Radial distortion can be estimated ahead of time to help with image alignment and mosaicking.

Uploaded by

Agent Tiger
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
34 views5 pages

CV Potential Questions

The document contains questions related to computer vision concepts like structure from motion, bundle adjustment, triangulation, fundamental matrix, and panoramic image processing. Bundle adjustment is now the standard method for structure from motion problems involving hundreds of weakly calibrated images and thousands of points. Bundle adjustment combined with gap closing can automatically assemble a panorama if the user takes overlapping images in sequence and specifies the start and end images. Professional panoramic photographers often use pan-tilt heads that make precise stops at specific rotation angles to aid in panorama stitching. Radial distortion can be estimated ahead of time to help with image alignment and mosaicking.

Uploaded by

Agent Tiger
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 5

CV POTENTIAL

QUESTIONS
1.The problem of determining a point’s 3D position from a set of corresponding image locations and
known camera positions is _.
a. Triangulation
b. Factorization
c. Bundle adjustment
d. Orthographic projection
2.The___ refers to the fact that motion estimation is highly ambiguous when the observation window
is very small.
a. panoptic segmentation
b. aperture problem
c. geometric reasoning
d. calibration error
3.One if the useful property of Fourier transforms is that convolution in the spatial domain
corresponds to __ in the Fourier domain
a. addition
b. subtraction
c. multiplication
b. division
4.Bundle adjustment is now the standard method of choice for most _ problems and is commonly
applied to problems with hundreds of weakly calibrated images and tens of thousands of points.
a. motion-from-structure
b. structure-from-motion
c. motion-from-motion
d. structure-from-structure
5.Which of the following can also be computed on the basis of line matches alone?
a. Trifocal Tensor
b. Tensor Flow
c. Stack overflow
d. both Tensor Flow and Stack overflow
6.__ technique can be used to estimate a series of rotation matrices and focal lengths, which can be
chained together to create large panoramas.
a. bundle adjustment
b. Parallax removal
c. Gap closing
d. Composting
7.Before we can register and align images, we need mathematical relationships that __ from one
image to another.
a. align an image
b. map pixel coordinates
c. rotate pixel
d. compare image
8.In ____ images are translated, optionally rotated and scaled.
a. panography
b. cryptography
c. photography
d. cartography
9.An alternative to using homography or 3D motions to align images is to first warp the images into _
and then use a pure translational model to align them.
a. Spherical coordinates
b. Cylindrical coordinates
c. Planar coordinates
d. Vector coordinates
10.If the user takes images in sequence so that each image overlaps its predecessor and also specifies
the first and last images to be stitched, bundle adjustment combined with the process of ___ can be
used to automatically assemble a panorama.
a. topology inference
b. mean difference
c. sum of squared difference
d. change in image
11.The fundamental matrix is given by ____.
12.Most modern cameras have __ pixels and an image center near the ____ of the image
a. round, center
b. rectangle, corner
c. square, middle
d. round, edge
13.Which one of the following is the point of intersection of the line joining the camera centers with
the image plane?
a. Epipole
b. Axis
c. Point of projection
d. Point at infinity
14.Fourier-based alignment relies on the fact that the Fourier transform of a shifted signal has the
same magnitude as the original signal, but a _____ phase
a. exponentially varying
b. linearly varying
c. unvarying
d. constant
15.Structure from motion is a ____in structure and motion.
a. Bipartite problem
b. Graph Coloring Problem
c. Travelling salesman problem
d. Normalized Cut
16.Triangulation is the converse of ___ problem.
a. Direct Linear Transform
b. Pose Estimation
c. 2D Motion Estimation
d. Rigid Transform
17.For converting a projective reconstruction into a metric one, _____ techniques have been
developed.
a. Orthographic Projection
b. Projection Matrix
c. Self-calibration
d. Epipolar
18.The normal vector perpendicular to the line can be expressed as a function of two angles using
______.
a. Spherical coordinates
b. cylindrical coordinates
c. Planar coordinates
d. Vector coordinates
19.Professional panoramic photographers often use pan-tilt heads that make it easy to control the tilt
and to stop at specific ___in the rotation angle.
a. error
b. bias
c. weight
d. detents
20.Radial distortion can be estimated ____.
a. ahead of time
b. Just in time
c. using alignment
d. by matching pixels
4 Marks:
UNIT IV:
1. What is the need for Projective Reconstruction?
2. Write short notes on the Bundle Adjustment techniques involved in accurately recovering structure
and motion.
3. Give the equation and briefly explain the components of Brightness Constancy Constraint equation.
4. Why does the Essential Matrix (E) change into the Fundamental Matrix (F)? Discuss in brief.
5. Give the equation and briefly explain the components of Schur complement.
6. What is self-calibration? What are the basic assumptions taken by any self-calibration method?

UNIT V:
1. Explain the Projection from 3D to cylindrical coordinate
2. Explain the process of Recognizing Panoramas.
3. What is parallax? How can it be removed?
4. Explain the Projection from 3D to Spherical coordinate
5. What are the various difficulties faced while Recognizing Panoramas?
6. What are the problems in producing the final stitched mosaic image? List the techniques used to
address the image stitching problem.
12 Marks:
UNIT IV:
1. What is Fourier-based Alignment? Explain it in detail with necessary examples.
2. Explain Hierarchical Motion Estimation in detail.
3. Briefly explain the following methods to establish a transitional alignment between two images or
image patches:
i) Spatially varying weights
ii) Bias and gain (exposure differences)
iii) Correlation
4. Examine the various Bundle Adjustment techniques involved in accurately recovering structure and
motion.

UNIT V:
1. Explain Parametric Motion Models in detail with necessary figures.
2. Explain the concept of Bundle Adjustment in detail.
3. Draw neat figures for "Projection from 3D to cylindrical and spherical coordinates" and explain the
same.
4. How do panoramas are recognized? Explain various methods and algorithms used for recognizing
panoramas

You might also like