Lect 5 Post-Processing and Display

Question 1

1. Name some applications that require a perceptually faithful display.

Answer 1

• Many applications that use electronic display devices require images to appear a certain way,

• e.g. medical imaging(make sure that data is displayed properly)
• aviation(e.g. simulating effect of copit window.. screen glare)
• visualisation(Archaeology)
• photography (Archaeology)
• predictive lighting(simulating scenes using ray tracing)
• realistic image synthesis.

Question 2

2. Why would similarity be desirable between image creation and end result?

Answer 2

To create a perceptually faithful display

• Similarity is desirable between the image as it was created and the resultant image that is viewed by the end-user.
• The user must be confident that the image they are viewing is faithful to the original – they require perceptual fidelity.

Question 3

3. What is a key problem with displaying images accross multiple devices?

Answer 3

• Images are often displayed on different monitors and in different locations from where they were created making the perceived image hard to keep consistent.

Question 4

4. What needs to be considered when displaying images if perceptual consistency is required?

Answer 4

Any point in an image should look the same regardless of changes in viewing location and display device.

It is necessary to be aware of any factors that might adversely influence the display medium.

Question 5

5. What assumption is made for perceived image consistancy?

Answer 5

• We have to make assumptions that the human visual system (HVS) is relatively consistent from person to person.
• Assuming that, and imagining that our image is correct, then accurately capturing or simulating light and colour result in a perceptually accurate image for most people.

Question 6

6. Name 3 areas where technology can let us down.

Answer 6

• Technology lets us down. We have problems with:

• Display limitations
• Viewing environment
• Colour constancy

Question 7

7. Why are computer monitors so poor at representing luminance?

Answer 7

• Computer monitors are limited in the range of luminances they can display. Most are not yet capable of producing anywhere near the range of light in the real-world.
• This means the realistic images we have carefully created aren’t being properly displayed.
• Humans can see starlight using peripheral vision.. cones.
• Also see snow, sunlight.. very bright..

Question 8

8. Describe CRT monitors.

Answer 8

• A colour Cathode Ray Tubes (CRT)uses three electron guns (referred to as red', green’ and `blue’ guns) which emit an electron beam.
• When a digital image is created it is stored as an array of values that represent an intensity(colour) of a particular part of that image. These values that are used to express colour actually specify the voltage that will be applied to each electron gun.
• The values are converted from digital to analogue, and video signals are produced, exciting the phosphors of the display and emitting light, which results in an image on screen.
• CRTs have a non-linear relationship between the input voltage and the light output.

Question 9

9 Describe the non-linear relationship between CRT input and output.

Answer 9

• The non-linear relationship between the CRT input and output approximates to a power law that is coincidentally close to the inverse of human luminance sensitivity (perceived brightness).

Question 10

10. How is gamma correction used in CRT?

Answer 10

• Since it is desirable for the displayed output on a CRT to be linear with brightness, gamma correction can be used to map luminance into a perceptually uniform domain.
• If images are not gamma encoded, they allocate too many bits to highlights that humans cannot differentiate, and too few bits to shadow values that humans are sensitive to and would require more bits to maintain the same visual quality.

Question 11

11. Describe LCD screens in relation to gamma.

Answer 11

• The relationship between signal voltage and intensity in an LCD device is very non-linear.
• Many LCD screens incorporate circuitry to mimic the transfer function of a CRT display device (with a gamma value of about 2.5)
• This means that although gamma is specifically aimed at correcting the non-linear transfer function of CRTs, it can often (but not always) be applied to images for display on LCD screens.

Question 12

12. Describe Gamma Correction

Answer 12

• For ‘correct’ gamma correction, the black level of the display device needs to be set to 256.

• Next, a gamma correction chart can be displayed and the device’s gamma value can be estimated (not appropriate for LCD screens).
• Once the gamma value is known, images can be pre-corrected before displaying.
• Often, proper correction is not carried out and instead a value of 2.2 is used. This boosts contrast (which users prefer).

Question 13

13. What is **Dynamic Range? **

Answer 13

• The ratio between maximum and minimum tonal values is known as the dynamic range.
• For an image, the dynamic range is the ratio between the lightest and darkest pixels.
• For a display, the dynamic range is the ratio of the maximum and minimum luminance that it is capable of emitting.

Question 14

14. Describe Dynamc Range in the real world.

Answer 14

• The real world has a much, much larger dynamic range than the two orders of magnitude common in current digital imaging: the sun at noon may be 100 million times lighter than starlight, for example.
• The Human Visual System (HVS) can adapt to lighting conditions that vary by nearly 10 orders of magnitude.

Question 15

15. Describe High Dynamic Range (HDR) images.

Answer 15

• HDR images store a depiction of the scene in a range of intensities commensurate with the real-world scene.
• These images may be rendered images or photographs.

Question 16

16. What is the difference between cameras that take HDR and those that do not.

Answer 16

• Non-HDR cameras take pictures at one exposure level with a limited contrast range. This results in the loss of detail in bright or dark areas of a picture, depending on whether the camera had a low or high exposure seeng.
• HDR compensates for this loss of detail by taking multiple pictures at different exposure levels and intelligently stitching them together to produce a picture that is representative in both dark and bright areas.

HDR is now on iOS 7.. mutiple pictures taken..

Question 17

17. How are digital photographs usually stored?

Answer 17

• Digital photographs are often encoded in a camera’s raw image format, because 8 bit JPEG encoding doesn’t offer enough values to allow fine transitions (and introduces undesirable effects due to the lossy compression).

Question 18

18. Describe HDR file formats.

Answer 18

• HDR images do not user integer values to represent the single color channels (e.g. 0..255 in an 8 bit per pixel interval for R, G and B) but instead use a floating point representation. Three of the most common file formats are as follows:

Question 19

19. Name 3 HDR file formats.

Answer 19

1. Radience RGBE
2. OpenEXR
3. **Floating point TIFF/PSD **

Question 20

20. Descibe Tone Mapping.

Answer 20

• Tone reproduction (also known as tone mapping) provides a method of mapping luminance (tonal) values in the real world to a displayable range.
• Tone reproduction is necessary to ensure that the wide range of light in a real-world scene is conveyed on a display with limited capabilities.

• A HDR image needs to be scaled in some way to be output on a display device that is only capable of outputting a low dynamic range.

Question 21

21. How does Tone Mapping compensate for **display limitations. **

Answer 21

• Tone mapping is a widely used method of mapping luminance values in the real world to a displayable range
• It aims to achieve a perceptual match between a real-world scene and a virtual scene by preserving aspects of an image such as contrast, brightness or fine detail.

Question 22

22. Linear scalling vs. **tone mapping. **

Answer 22

• A straighforward linear scaling between the original high dynamic range(HDR)data and the display is not the best as many - if not all-details can be lost. The mapping must be tailored in some non-linear way.
• Linear scaling of HDR data will cause almost all details to be lost, (top image). Here, the light bulb is mapped to a few white pixels and the remainder of the image is black. Tone mapping recovers detail in both light and dark areas as well as all areas in between (bottom image).

Question 23

23. Describe the two types of Tone Mapping.

Answer 23

• Two types of tone reproduction operators can be used: spatially uniform - also known as single-scale or global, and spatially varying - also known as multi- scale or local.
• Global operators apply the same transformation to every pixel. It may depend upon the contents of the image as a whole, but the same transformation is applied to every pixel.
• Conversely, local operators apply a different scale to different parts of an image.

Question 24

24. Discuss Variation in Tone Mapping algorithms.

Answer 24

• A number of tone mapping algorithms have been developed, with each generally addressing a specific aspect such as brightness preservation or contrast preservation.
• Some are concerned with achieving perceptual fidelity with a real-world scene and mimic aspects of the human visual system. Others concentrate on producing a subjective best image that is pleasing to the eye.

Question 25

25. How does ambient light cause viewing problems?

Answer 25

• Viewing conditions affect our perception of an image.
• Ambient light in room causes displayed images to appear faded.
• Goal: we want images to look the same even if they are viewed under different conditions.

Question 26

26. How do we compensate for ambient light in the viewing environment.

Answer 26

• Compensation can be made to restore contrast to the image
• Metadata about the creation and expected viewing conditions can be included
• Technology (e.g. monitors with ambient light sensors) can be employed

Question 27

27. What is colour constancy.

Answer 27

• Colour constancy refers to the way in which our visual system interprets colours as unchanging, despite changes in illumination.
• When we simulate the colour of the light source and then produce results that show perceptual differences, are these differences actually valid?

Question 28

28. How do we conpensate for **colour constancy. **

Answer 28

• Provide information about the scene illumination (e.g. technical metadata describing spectral values and material properties).
• Look to colour science, colour vision and optometry research