web of technologies

94

Click here to load reader

Upload: dynamis-

Post on 14-Jul-2015

42.969 views

Category:

Technology


17 download

TRANSCRIPT

Page 1: Web of Technologies
Page 3: Web of Technologies

OOHH22

Page 5: Web of Technologies
Page 6: Web of Technologies
Page 8: Web of Technologies
Page 9: Web of Technologies

TT ff PP

pp.. FF dd MM

oo tt

Page 10: Web of Technologies

xxzz ww

nn HHssjjCC

Page 11: Web of Technologies
Page 12: Web of Technologies

WEB MOBILE FIREFOX OS

Page 13: Web of Technologies
Page 14: Web of Technologies
Page 15: Web of Technologies

29 ヶ国

15 キャリア

Page 16: Web of Technologies
Page 17: Web of Technologies
Page 18: Web of Technologies
Page 19: Web of Technologies

WEB OF THINGS

Page 20: Web of Technologies
Page 27: Web of Technologies
Page 29: Web of Technologies
Page 33: Web of Technologies

O S W F

I

 L E B T

light2light1IrDAmotion sensorSwitch

Page 34: Web of Technologies
Page 35: Web of Technologies
Page 37: Web of Technologies
Page 39: Web of Technologies
Page 40: Web of Technologies
Page 41: Web of Technologies
Page 42: Web of Technologies

vvkk

JJ

Page 43: Web of Technologies
Page 44: Web of Technologies

SerialUDB Diagram Socket

PrintUSB

WebCL WebGL2

Page 45: Web of Technologies

DeviceIndexedDBUSB File Reading

WebSocket Over Apps

Page 46: Web of Technologies

ccbbhhaa HH

11TT ffcc

CCrr zzllFFrr 11

Page 47: Web of Technologies
Page 48: Web of Technologies
Page 49: Web of Technologies
Page 52: Web of Technologies
Page 54: Web of Technologies
Page 56: Web of Technologies

So, the Project FoxEye came to the world.

Page 57: Web of Technologies
Page 58: Web of Technologies
Page 59: Web of Technologies

navigator.getUserMedia( {video:true, audio:false}, function(localMediaStream) { var videoTracks = localMediaStream.getVideoTracks(); var track = videoTracks[0]; var myWorker = new Worker("textRec.js"); track.addWorkerMonitor(myWorker)); myWorker.onmessage = function (oEvent) { console.log("Worker recognized: " + oEvent.data); }; var elem=document.getElementById('VIDEO'); elem.mozSrcObject = dest.stream; elem.play(); }, null);

Page 60: Web of Technologies

var textDetector = WebImage.createTextDetector(img.width, img.height) onvideoprocess = function (event) { var img = event.inputFrame; // テキスト認識処理を実行 var words= textDetector.findText(img); var recognizedText; for (var ix=0; ix<words.length; ix++) { recognizedText = recognizedText + words[ix] + " "; } postMessage(recognizedText); };

Page 61: Web of Technologies

Use cases(Amazon FireFly)In our case, we can use Yahoo Service, like Yahoo Movie, Yahoo Shopping, Yahoo Search...

Page 62: Web of Technologies

Use cases(Text selection in Image, browser)

http://projectnaptha.com/

Page 63: Web of Technologies

Use cases(Augmented Reality)

Page 64: Web of Technologies

Use cases(Face in)

Page 65: Web of Technologies

Use cases(Camera Panorama, HDR)

Page 66: Web of Technologies

WWii SShhii

Page 68: Web of Technologies
Page 69: Web of Technologies
Page 70: Web of Technologies
Page 75: Web of Technologies
Page 76: Web of Technologies
Page 77: Web of Technologies
Page 81: Web of Technologies
Page 82: Web of Technologies
Page 83: Web of Technologies

17

Emscripten now targets SIMD.JS

Emscripten generates SIMD.JS from C++ SIMD intrinsics & auto-vectorized code

Near-native SIMD.JS speedup

C/C++ JavaScript

1.002.03

7.188.13

02468

10

Speedup over Scalar JS

Scalar JS Scalar C++

SIMD JS SIMD C++

11

SIMD – Single Instruction, Multiple Data

SIMD operations deliver great performance & power efficiency

Scalar Operation

Cx

Cy

Cz

Cw

=

=

=

=

Ax

Ay

Az

Aw

Bx

By

Bz

Bw

+

+

+

+

Cx

Cy

Cz

Cw

Ax

Ay

Az

Aw

Bx

By

Bz

Bw

+ =

SIMD Operation of Vector Length 4

Intel® Architecture currently has SIMD operations of vector length 4, 8, 16

Page 85: Web of Technologies

11

SIMD – Single Instruction, Multiple Data

SIMD operations deliver great performance & power efficiency

Scalar Operation

Cx

Cy

Cz

Cw

=

=

=

=

Ax

Ay

Az

Aw

Bx

By

Bz

Bw

+

+

+

+

Cx

Cy

Cz

Cw

Ax

Ay

Az

Aw

Bx

By

Bz

Bw

+ =

SIMD Operation of Vector Length 4

Intel® Architecture currently has SIMD operations of vector length 4, 8, 16

Page 86: Web of Technologies

13

SIMD.JS – The API

† Initial support for float32x4 and int32x4

Our SIMD prototype delivers 3x~4x Mandelbrot speedup†

Our Firefox* Prototype

Page 87: Web of Technologies

15

SIMD Speedups on our Chromium* Prototype

3.2 3.6 3.8 3.94.6

5.0

6.0

9.5

3.23.8

3.4

6.16.5

5.05.6

11.8

6.8

3.12.7

4.5 4.2 3.8

5.4

9.3

0

2

4

6

8

10

12

14

Transpose4x4 AOBench Mandelbrot MatrixMultiplication VertexTransform Average ShiftRows Matrix4x4Inverse

SIMD x-times faster than non-SIMD

���3rd Generation Intel® Core™ i7 processor (3667U)@ 2.00 GHz, 32-bit, Ubuntu* 133rd Generation Intel® Core™ i7 processor (3667U)@ 2.00 GHz, 64-bit, Ubuntu* 13 Intel® Atom™ processor Z3770 @ 1.46GHz, Android* 4.4

Excellent early results while still focused on functionality

Theoretical speedup limit is 4

SIMD.JS benchmarks: https://github.com/johnmccutchan/ecmascript_simd/tree/master/src/benchmarks

Page 88: Web of Technologies

17

Emscripten now targets SIMD.JS

Emscripten generates SIMD.JS from C++ SIMD intrinsics & auto-vectorized code

Near-native SIMD.JS speedup

C/C++ JavaScript

1.002.03

7.188.13

02468

10

Speedup over Scalar JS

Scalar JS Scalar C++

SIMD JS SIMD C++

Page 89: Web of Technologies
Page 90: Web of Technologies

PP ee

mm RRHHssjj

HH ccHH

Page 91: Web of Technologies
Page 92: Web of Technologies

WEB OF THINGS

Page 93: Web of Technologies

!!

Page 94: Web of Technologies