Question Bank

CSC503: Computer Vision

1. …… data doesn’t matter as long as there are enough features remaining to agree on a
good model.
a) Missing b) reflected c) Refracted d) transmitted
2. …….. can detect multiple instances of a model in a single pass. .
a) Hough Transform b) FFT c) Fourier Transform d) DWT
3. ……….. can be used to initialize fitting for more complex models. .
a) Affine Transformations b) Hough Transform c) FT d) IDWT
4. Each match gives us two linearly independent equations: need at least …… to solve
for the transformation parameters.
a) one b) three c) four d) five
5. We need …. geometry because recovery of structure from one image is inherently
ambiguous.
a) single-view b) line-view c) multi-view d) geometrical correct view
6. …….. axis is a line from the camera center perpendicular to the image plane. .
a) Main b) Principal c) Epipolar d) Perpendicular
7. We can write the epipolar ………. in terms of unknown normalized coordinates. .
a) constraint b) base line c) pole d) plane
8. Given a calibrated binocular stereo pair, fuse it to produce a……….
a) original image b) 3D image c) 2D image d) depth image
9. In plane sweep stereo, the sampling of the scene still depends on the ………. view.
a) reference b) original c) camera d) stereo
10. If we scale the entire scene by some factor k and, at the same time, scale the camera
matrices by the factor of 1/k, the projections of the scene points in the
image…………
a) remain exactly the same b) changes randomly c) convert to 3D d) convert to
2D
11. Rays passing through the center of lens are not ……..
a) deviated b) reflected c) Refracted d) transmitted
12. All ….. rays converge to one point on a plane located at the focal length f.
a) parallel b) perpendicular c) diverged d) intersect
13. A …… aperture increases the range in which the object is approximately in focus.
a) smaller b) large c) circular d) irregular
14. ….. the lens with respect to the image plane allows to choose an arbitrary plane of
focus.
a) focused b) Tilting c) concentric d) circular
15. Lens has different refractive indices for different wavelengths: causes ….. fringing.
a) refraction b) reflection c) color d) image
16. Deviations are most noticeable for rays that pass through the ….. of the lens.
a) center b) edges c) middle d) parallel
17. …… irradiance is linearly related to scene radiance.
a) Image b) system c) state d) area
18. In the most general case, a single incoming ray could be reflected in ……. directions
a) only one b) straight c) curve d) all
19. Color is the result of interaction between physical light in the environment and our
…….
a) visual system b) input device c) camera d) sensor
20. An …… is a place of rapid change in the image intensity function
a) edge b) color c) region d) pattern
21. The perspective distortion is not due to …… flaws.
a) camera b) focal length c) lens d) illumination and reflection
22. Perspective Projection is a matrix …….. using homogeneous coordinates.
a) multiplication b) addition c) subtraction d) summation
23. By analogy with angle (in radians), the …… subtended by a region at a point is the
area projected on a unit sphere centered at that point.
a) solid square b) solid circle c) solid triangle d) solid angle
24. …. is the result of interaction between physical light in the environment and our
visual system.
a) Color b) Image c) Picture d) Photograph
25. Image …….. results in pixels that look very different from their neighbors.
a) intensity b) face c) noise d) color
26. The same feature can be found in several images despite … and photometric
transformations.
a) geometric b) mathematical c) visual d) camera
27. Higher-level and more compact representation of the features in the image is
calculated by grouping ……. features according to a simple model. .
a) basic b) high level c) single d) multiple
28. …….. doesn’t matter as long as there are enough features remaining to agree on a
good model.
a) Missing data b) Simple data c) ordinary data d) Extracted data
29. In Alignment: fitting a model is a transformation between pairs of features (matches)
in two…….
a) features b) images c) points d) data points
30. Points move along lines radiating from …… will be “Focus of expansion”.
a) Focal Length b) Epipolar Point c) Epipolar Line d) Epipoles e
31. In Camera coordinate system: the origin is start at the ….. point.
a) last b) first c) focal d) principal
32. Given a calibrated binocular stereo pair, fuse it to produce a ….. of image.
a) height b) depth c) width d) length
33. In multiple base line stereo for larger baselines, must search ……. area in second
image.
a) larger b) small c) depth d) detail
34. With no constraints on the camera calibration matrix or on the ….. , we get projective
reconstruction.
 a) scene b) fundamental matrix c) both a and b d) none
35. Invariant to similarity transformations is computed from ……. points.
a) five b) three c) four d) two
36. Generative method can be learned using images from just a ….. category.
a) three b) two c) multiple d) single
37. Compute an updated estimate of the state from ………. and measurements.
a) last location b) current target c) current position d) prediction
38. For computational efficiency, we should try to spend as little time as possible on the
non-face ……….
a) windows b) image c) both a and b d) none of this
39. In Discriminative methods learn a decision rule (classifier) and assigning bag-of-
features representations of images to …… classes.
a) different b) same c) exactly similar d) image
40. The Naïve Bayes model assume that each feature is conditionally ……… given in the
class.
a) same b) dependent c) independent d) different