26 3 Stereo shape constancy with active and passive movement 1175068 2015 2 27
3 3D VR 3D 2 2 i
ii
Abstract Stereo shape constancy with active and passive movement Katsuya Okubo It will become more usual for observers to move while observing 3D images as virtual reality becomes widely used. The stereoscopic 3D perception is achieved by presenting images with binocular disparity and the binocular disparity changes inversely with the square of the viewing distance. Thus, even if we see the same object, the retinal disparity changes according to the viewing distance. Although previous study investigated the depth scaling when the target object is moved in the direction of depth in near distance, little is known how we perceive the depth of disparity-defined objects in far distance when we passively see the objects coming closer or actively move toward the objects. The purpose of this study is to investigate the shape constancy from binocular disparity with object motion or self-motion at far distance in which the disparity becomes small examining the point of subjective equality of depth within object. In addition, to examine the effect of a continuous change of disparity in real time during the objects or the observers movement, we set the condition which showed the object with continuous change of disparity and the condition which showed only two static frames at the timing of the start and end of the movement. The results showed that in any conditions, the depth of the object which appeared to be constant was smaller than the theoretical value at a relatively shorter distance. These results mean that the flatter object in nearer distance seemed to be the same depth at a far distance. These results might come from the underestimation of the movement distance, or increase of binocular disparity. As correlation analysis between perceived movement and perceived depth revealed no iii
correlation among them, it is plausible that the result may be caused by the increase of disparity. Also this result suggests that the continuous and gradual change of disparity does not have a significant effect on accurate depth scaling. key words Binocular disparity,shape constancy, viewing distance, depth, movement, random dot stereogram iv
1 1 1.1...................................... 1 1.2................................... 2 1.3................................. 4 1.4............................ 4 1.5...................................... 5 2 7 2.1.................................... 7 2.2...................................... 7 2.3...................................... 10 2.4.................................... 12 2.4.1 1 2....................... 12 2.4.2 1......................... 13 2.4.3 2 2....................... 15 2.4.4 2......................... 16 2.4.5 3 2....................... 18 2.4.6 3......................... 18 3 20 3.1 1..................................... 20 3.1.1 2............................ 20 3.1.2............................. 24 3.1.3............................. 28 v
3.2 2..................................... 29 3.2.1 2............................ 29 3.2.2............................. 33 3.2.3............................. 37 3.2.4............... 37 3.2.5 3................................. 39 3.2.6 2............................ 39 3.2.7............................. 43 3.2.8............................. 47 3.3...................................... 47 4 49 50 51 vi
1.1 [6].................................. 2 1.2 [10]......................... 3 1.3......................... 5 2.1 3DDLP (InFocus DepthQ HD)............... 8 2.2 (KIC ES-120W)................. 8 2.3 CRT EIZO FlexScan................. 8 2.4 Logicool K270...................... 8 2.5 NuVision 60GX................. 9 2.6 3D NuVision 60GX-T50.................. 9 2.7 3 POLHEMUS FASTRAK........ 9 2.8...................................... 9 2.9................................. 11 2.10................................... 11 2.11................................... 11 2.12..................................... 11 2.13 1..................................... 14 2.14 2..................................... 17 2.15 3..................................... 19 3.1 1 2......................... 21 3.2 A 1 2................... 21 3.3 B 1 2................... 22 3.4 C 1 2................... 22 vii
3.5 D 1 2................... 23 3.6 E 1 2................... 23 3.7 1.......................... 25 3.8 A 1.................... 25 3.9 B 1.................... 26 3.10 C 1.................... 26 3.11 D 1.................... 27 3.12 E 1.................... 27 3.13.................... 28 3.14 2 2......................... 29 3.15 A 2 2................... 30 3.16 B 2 2................... 30 3.17 C 2 2................... 31 3.18 D 2 2................... 31 3.19 E 2 2................... 32 3.20 2.......................... 33 3.21 A 2.................... 34 3.22 B 2.................... 34 3.23 C 2.................... 35 3.24 D 2.................... 35 3.25 E 2.................... 36 3.26.................... 37 3.27 2.......................... 38 3.28 3 2......................... 39 3.29 A 3 2................... 40 3.30 B 3 2................... 40 viii
3.31 C 3 2................... 41 3.32 D 3 2................... 41 3.33 E 3 2................... 42 3.34 3.......................... 43 3.35 A 3.................... 44 3.36 B 3.................... 44 3.37 C 3.................... 45 3.38 D 3.................... 45 3.39 E 3.................... 46 3.40.................... 47 ix
1 1.1 VR 3D 3D 3D 3 3D 3D [1][2] [3][4][5] 1
1.2 1.2 1.1 [6] 6 cm 10 [7] [8] 3D [9] 2
1.2 1.2 [10] 1.2 F Q 2 I 6 cm F 1 D F Q D 1.2 F Q Θ 1 Θ 2 Θ 1 Θ 2 Θ Θ = Θ 1 Θ 2 = α 2 α 1 ( a D D) a D a D D 2 = 206265 a D D 2 3
1.3 D 2 [11] 1.3 6 12 18 4 m 4 m [12] 1.4 [13] 2 3 4
1.5 1.3 1.3 2 [3] 3 3 [14] 1.5 5
1.5 2 6
2 2.1 20 5 2.2 MATLAB+Psychtoolbox 1 2 3DDLP (InFocus DepthQ HD, 120 Hz) (KIC ES-120W) 3 21 CRT EIZO FlexScan T961 1024 768pixel 120Hz 3D NuVision 60GX 3D NuVision 60GX-T50 Logicool K270 2 3 POLHEMUS FASTRAK 3 7
2.2 2.1 3DDLP (InFocus DepthQ HD) 2.2 (KIC ES-120W) 2.3 CRT EIZO FlexScan 2.4 Logicool K270 8
2.2 2.5 NuVi- 2.6 3D NuVision sion 60GX 60GX-T50 2.7 3 POLHEMUS FASTRAK 2.8 9
2.3 2.3 1 2 18 cm 36 cm 3 6 cm 12 cm 1 2 270 cm 3.82 deg 7.63 deg 180 cm 5.72 deg 11.42 deg 3 90 cm 3.82 deg 7.63 deg 60 cm 5.72 deg 11.42 deg 2000 1 2 270 cm 9 cm 180 cm 4.5 6.0 7.5 9.0 10.5 12.0 13.5 cm 7 180 cm 7 3 90 cm 3 cm 60 cm 1.5 2.0 2.5 3.0 3.5 4.0 4.5 cm 7 60 cm 7 2.2 2.3 2 1 2 1.69 cm 1.69 cm 3 1.125cm 1.125 cm 10
2.3 2.9 2.10 2.11 2.12 11
2.4 2.4 2.4.1 1 2 270 cm 5 270 cm 9 cm 1.5 s 1.5 s 270 cm 180 cm 180 cm 90 cm 1.5 s 180 cm 7 1.5 s 2 6 4 1 7 10 3 210 12
2.4 2.4.2 1 270 cm 5 270 cm 9 cm 1.5 s 180 cm 7 180 cm 1.5 s 1.5 s 2 6 4 1 7 10 3 210 13
2.4 2.13 1 14
2.4 2.4.3 2 2 3 270 cm 5 270 cm 9 cm 1.5 s 180 cm 180 cm 270 cm 100% 180 cm 0% 180 cm 1.5 s 180 cm 7 2 6 4 1 7 10 3 210 15
2.4 2.4.4 2 3 270 cm 5 270 cm 9 cm 1.5 s 180 cm 3 180 cm 270 cm 100% 180 cm 0% 180 cm 180 cm 1.5 s 2 6 4 1 7 10 3 210 16
2.4 2.14 2 17
2.4 2.4.5 3 2 90 cm 5 90 cm 3 cm 1.5 s 1.5 s 90 cm 60 cm 60 cm 30 cm 1.5 s 60 cm 7 1.5 s 2 6 4 1 7 10 3 210 2.4.6 3 90 cm 5 90 cm 3 cm 1.5 s 180 cm 7 60 cm 1.5 s 1.5 s 2 6 18
2.4 4 1 7 10 3 210 2.15 3 19
3 3.1 1 3.1.1 2 5 3.1 3.2 3.6 180 cm 0.5 8.3 cm 9 cm t 4 = 4.82 p <.05 20
3.1 1 3.1 1 2 3.2 A 1 2 21
3.1 1 3.3 B 1 2 3.4 C 1 2 22
3.1 1 3.5 D 1 2 3.6 E 1 2 23
3.1 1 3.1 A 7.9 B 8.3 C 8.2 D 8.5 E 8.8 3.1.2 5 3.7 3.8 3.12 180 cm 0.5 270 cm 9 cm 8.4 cm 9 cm t 4 = 2.19 p = 0.07 24
3.1 1 3.7 1 3.8 A 1 25
3.1 1 3.9 B 1 3.10 C 1 26
3.1 1 3.11 D 1 3.12 E 1 27
3.1 1 3.2 A 7.9 B 7.7 C 8.8 D 8.2 E 9.3 3.1.3 1 3.13 1 2 3.13 28
3.2 2 3.2 2 3.2.1 2 5 3.34 3.35 3.39 180 cm 0.5 8.7 cm 9 cm t 4 = -1.46 p = 0.19 3.14 2 2 29
3.2 2 3.15 A 2 2 3.16 B 2 2 30
3.2 2 3.17 C 2 2 3.18 D 2 2 31
3.2 2 3.19 E 2 2 3.3 A 8.3 B 8.0 C 8.9 D 8.9 E 9.2 32
3.2 2 3.2.2 5 3.20 3.21 3.25 180 cm 0.5 8.0 cm 9 cm t 4 = -6.58 p <.05 3.20 2 33
3.2 2 3.21 A 2 3.22 B 2 34
3.2 2 3.23 C 2 3.24 D 2 35
3.2 2 3.25 E 2 3.4 A 7.7 B 8.1 C 8.2 D 8.4 E 7.7 36
3.2 2 3.2.3 2 3.40 2 t 4 = 2.53 p <.05 3.26 3.2.4 5 2 3.7 3 2 3.27 2 37
3.2 2 3.5 cm A 80 B 80 C 70 D 100 E 100 3.27 2 38
3.2 2 3.2.5 3 3.2.6 2 5 3.34 3.35 3.39 60 cm 0.5 2.7 cm 3 cm t 4 = 6.18 p <.01 3.28 3 2 39
3.2 2 3.29 A 3 2 3.30 B 3 2 40
3.2 2 3.31 C 3 2 3.32 D 3 2 41
3.2 2 3.33 E 3 2 3.6 A 2.8 B 2.7 C 2.7 D 2.8 E 2.5 42
3.2 2 3.2.7 5 3.34 3.35 3.39 60 cm 0.5 2.7 cm 3 cm t 4 = 5.76 p <.01 3.34 3 43
3.2 2 3.35 A 3 3.36 B 3 44
3.2 2 3.37 C 3 3.38 D 3 45
3.2 2 3.39 E 3 3.7 A 2.9 B 2.7 C 2.6 D 2.6 E 2.5 46
3.3 3.2.8 3 3.40 2 p <.05 3.40 3.3 1 2 2 2 3 47
3.3 2 2 1 2 3D 3D 48
4 3D 3D 3D 49
50
[1] Mayhew JE Longuet-Higgins HC A computational model of binocular depth perseption 1982 Nature 297 pp.376-378 [2] Howard IP Rogers BJ Binocular Vision and Stereopsis 1995 Oxford Univ Press [3] P Scarfe P B Hibbard Disparity-defined objects moving in depth do not elicit three-dimensional shape constancy 2006 Vision Res 46 10 1599-610 [4] Shigemasu H Okubo K Yan P Shape constancy from binocular disparity with self-motion in depth 2013 Perception 42 ECVP Abstract Supplement pp.117 [5] Yan Pengfei 2014 Vision 26 1 pp.59 [6] Gibson J J The perception of the visual world 1950 Houghton Miffilin [7] 2008 [8] Howard IP Depth perception 2002 Seeing in Depth vol1 I Porteous [9] II pp 84-85 2007 [10] Graham C H Ed Vision and visual perception 1965 Wiley [11] Ono H and Commerford J Stereoscopic depth constancy 1977 Epstein W Stability and constancy in visual perception Wiley-Interscience Publication New York [12] 2001 Vision 13 pp 163-180 [13] Holway AH Boring EG Determinants of apparent visual size with distance variant American Journal of Psychology 1941 54 pp21-37 [14] 51
2004 [15] 2004 60(4) pp 455-459 [16] 2005 Vol 10 No 2 pp 249-256 [17] - - 2000 [18] 2000 [19] 2011 52