29 23 23115148
i 1 1 2 3 2.1..................................... 3 2.2.................................. 6 2.2.1............................... 6 2.2.2.................................. 8 3 10 3.1........................................ 10 3.2.......................................... 12 3.3..................................... 15 3.3.1....................... 15 3.3.2.......................... 17 4 18 19 20
1 1,. ( ). [1],. [2],.,, Albert, 93 [3].,. [4],, [5]., [6][7].,,,,.,., [8]. [9]
1 2, 700,...,,.,,..,,.,,.
3 2,,.,..,,.,,,.,. 2.1,,,., OpenCV, ( 2.1). (2.2.1, 2.2.2, OpenCV. ),,. w e, h e, (x, y).
2 4 weight, r, rg 2.1, 2.2. ( x w ) ( 2 e + y h ) 2 e r = 2 2 ( we ) ( ) 2 2 (2.1) he + 2 2 rg = r weight (2.2) r, (x, y),. rg r. r, rg (vx, vy) 2.3, 2.4. vx = (x w e) rg r vy = (y w e) rg r 2.2, 2.3. + w e + 0.5 (2.3) + h e + 0.5 (2.4)
第2章 テレビ会議システムにおける瞳の拡大表示 図 2.1: 両目の認識 図 2.2: 元の画像 図 2.3: 瞳の拡大 5
2 6 2.2,,.,.,.,,.,.,. [10].,.,,. 2.2.1.. t w f (t) t D 1 w f (t D 1 ). D 1. t D 1, t, c 1 ( 2.5, 2.4). c 1 < w f (t) w f (t D 1 ) (2.5)
2 7 2.4:
2 8 2.2.2, OpenCV,. P (t). D 2 P (t) P ave (t)..,,,. P (t),. t w f (t) w m (t) P (t) = w m (t)/w f (t). D 2 P (t) P ave (t) 2.6. P ave (t) = t k=t D 2 P (k) D 2 (2.6), P ave (t) P (t) c 2, ( 2.7, 2.5). c 2 < P (t) P ave (2.7)
2 9 2.5:
10 3,,,.,. 3.1 8 ( 7, 1 ) 4 2. 3, 1 4. 4, 2.,.,,. 5, 3. 3.1 (, Exp Eye, Video ).,,,. 2,...
3 11,.,. 1 Exp Eye Video 2 Video Exp Eye - ( ) 3 Exp Eye Video 4 Video Exp Eye - ( ) 3.1:
3 12 3.2 Skype[11]. YAMAHA PJP-10UR [12]. BUFFALO BSHSUH13 [13]. PC, Skype PC (CPU: Inter Core i5-6500 3.20GHz RAM:8.0GB). BUFFALO USB BSW20K04, PC (CPU: Inter Core i5-7200u 2.50GHz RAM:4.0G). PC 2 C++., D 1, D 2, c 1, c 2 3.2. weight 1.15 ( 3.2). 3.1:
第3章 評価実験 13 値 瞳を拡大表示する時間 1 秒間 指定時間 D1 0.27 秒 指定時間 D2 0.53 秒 閾値 c1 1.13 閾値 c2 1.08 表 3.2: 各パラメータの設定 図 3.2: 拡大処理後に表示される瞳の大きさ 図 3.3: 現地会議室の様子
第3章 評価実験 14 図 3.4: 遠隔会議室の様子
3 15 3.3 3.3.1, 3.3, 3.4., n n, n. 1 5 5,. 3.3 1, 2, 1.17, 1.58, 3.25, 2.75.,,.. 3,. 3.4 1, 2,., t p 0.18, 0.76,. 2 t p 0.0004,.,.,,,,. Video Exp Eye 1, 1.17 3.25. 2, 1.58 2.75. 3. 4.67 3.83 3.3:
3 16 Video Exp Eye 1, 2.5 3. 2, 2.75 3. 3.4:
3 17 3.3.2,.,., 3.1.,.,,. 3.5.,.... Video Exp Eye 40 50 9 13 ( )/( ) 0.23 0.26 3.5:
18 4,,.,,.,,.,,,.,,., (, D 1, D 2 c 1, c 2, weight)..,.,..
19
20 [1] Ellen A.Isaacs, John C.Tang, What Video Can and Can t Do for Collaboration : A case study, MULTIMEDIA 93 Proceedings of the first ACM international conference on Multimedia, 199-206, 1993. [2] W, :, ( ), ( ), 1981. [3],, ( ), ( ), 1995. [4],,,, Vol.58, No.1, 166-75, 2017. [5],,,, Vol.56, No.1, 126-136, 2015. [6],,, :,, Vol.57, No.1, 209-217, 2016. [7],,,, Vol.14, No.4, 497-504, 2015. [8],,,,,,,,,2002. [9] Ryo Oda, Ryota Ichihashi, The Watching Eyes Effect on Charitable Donation is Boosted by Fewer People in the Vicinity, Letters on Evolutionary Behavioral Sciense, Vol.7, No.2, 2016.
21 [10],,,,, ( ),Vol25, No.5, 889-900, 2013. [11] Microsoft Skype, https://www.skype.com/ja/. [12] YAMAHA PJP-10UR, https://sound-solution.yamaha.com/products/uc/pjp- 10ur/index. [13] BUFFALO BSHSUH13, http://buffalo.jp/product/multimedia/headset/ bshsuh13/. [14] BUFFALO BSW20k04, http://buffalo.jp/products/catalog/supply/ multimedia/webcamera/200/bsw20k04/.