Report itu-r bt. 2293-0 (11/2013)



Download 3.73 Mb.
Page12/14
Date conversion08.07.2018
Size3.73 Mb.
1   ...   6   7   8   9   10   11   12   13   14

2 Subjective measurement of visual discomfort induced by disparity characteristics

2.1 Visual stimulus


In order to determine the limit and the range of the allowable binocular disparity to guarantee the safety of 3DTV watching, the visual discomfort level should be assessed while changing the level of binocular disparity with a standard stimulus. For this study, video content with five different levels of standard stimuli were designed. As shown in Table 16, a total of 45 visual stimuli were generated (15 stimuli for disparity and30 stimuli for depth motion).

Video clips with different disparity levels were created by controlling the disparity precisely in a production studio. Positive disparity was set to the maximum 4.5% (86 pixels) in the positive direction, negative disparity was set to the maximum 4.0% (72 pixels) in the negative direction, and the depth budget was set to the maximum 4.5% (86 pixels) by assigning the positive and negative portions equally. Two types of depth motion, when the motion appeared just once or repeatedly, were implemented. Additionally, the change of disparity was set to 2%, 4%, and 6% all starting from 0%, and they were presented in terms of 5-level stimuli depending on the depth motion of the object.

TABLE 16

Attributes of visual stimulus used in the experiments





Depth position

Visual stimulus (disparity range)

Disparity

Positive

0.5%, 1.5%, 2.5%, 3.5%, 4.5%

Negative

0.5%, 1.2%, 2.0%, 3.0%, 4.0%

Depth budget

1.0%, 2.0%, 4.0%, 6.0%, 8.0%

Depth motion

Loop (negative)

Depth motion level

0%~2%, 0%~4%, 0%~6%

Standard stimulus duration

0.5 s, 1 s, 2 s, 4 s, 6 s

Stimulus presentation duration

15 s, 15 s, 15 s, 15 s, 15 s

Once (negative)

Depth motion level

0%~2%, 0%~4%, 0%~6%

Standard stimulus duration

0.5 s, 1 s, 2 s, 4 s, 6 s

Stimulus presentation duration

3.5 s, 4 s, 5 s, 7 s, 9 s

Figure 31



Disparity images used in the experiments

Stimulus

Positive





Camera

Sony HDR-1500R

Camera
movement

No

Rig

TS-2 (3Ality)

Characteristics

Fixed at negative 0%

Subject

movement

Yes

Image






Near distance (m)

Far distance (m)

Focal length (mm)

1.58 m

6.3 m

7.80 mm

IOD (%)

Convergence Point
(‘fit “inch)


Focus Point (‘fit “inch)

0.0%

6’ 24“
(Shoulder of the female subject)

6’ 43“
(Face of the female subject)



(a) Positive disparity images


A woman is sitting down at a coffee shop. A waiter comes with a cup of coffee, places it on the table, and goes back



Stimulus

Negative





Camera

Sony HDR-1500R

Camera

movement

No

Rig

TS-2 (3Ality)

Characteristics

Fixed at positive 0%

Subject

movement

Yes

Image






Near distance (m)

Far distance (m)

Focal length (mm)

1.6

6.2

7.80

IOD (%)

Convergence point (‘fit “inch)

Focus point (‘fit “inch)

0.0%

22’ 28“(Back wall)

6’ 26“

(Face of the female subject)



(b) Negative disparity images

A woman is sitting down at a coffee shop. A man sits down in front of her, and starts friendly conversation




Stimulus

Depth Budget





Camera

Sony HDR-1500R

Camera movement

No

Rig

TS-2 (3Ality)

Characteristics




Subject

movement

Yes

Image






Near distance (m)

Far distance (m)

Focal length(mm)

1

8.4

7.80

IOD (%)

Convergence point (‘fit “inch)

Focus Point (‘fit “inch)

0

7’ 70“(Front of the desk)

12’ 15“

(Face of the female subject)



(c) Depth budget images

After commentary at the desk, a female announcer comes forth and presents a final greeting



Higher resolution 1 920 × 1 080 side by side videos (mov(h.264)) of these sequences can be downloaded from the ITU site via this link http://www.itu.int/oth/R0A07000034/en.

Figure 32



Motion depth images used in the experiments

(a) Depth motion (loop) (b) Depth motion(once)


1   ...   6   7   8   9   10   11   12   13   14


The database is protected by copyright ©dentisty.org 2016
send message

    Main page