Upgrade to Pro
— share decks privately, control downloads, hide ads and more …
Speaker Deck
Sign up for free
Menu
Search
Features
All features
Private URLs
Password Protection
Custom URLS
Scheduled publishing
Remove Branding
Restrict embedding
Deck Collections
Notes
Features
All features
Private URLs
Password Protection
Custom URLS
Scheduled publishing
Remove Branding
Restrict embedding
Deck Collections
Notes
Explore
Featured decks
Featured speakers
Programming
Technology
Storyboards
Explore
Featured decks
Featured speakers
Programming
Technology
Storyboards
Pricing
Search
Sign in
Sign up for free
openCVで顔をごにょごにょする#TechLunch
Search
Livesense Inc.
PRO
April 21, 2014
Technology
140
0
Share
Embed
Copy iframe code
Copy JS code
Copy link
Start on current slide
openCVで顔をごにょごにょする#TechLunch
openCVで顔をごにょごにょする
2013/02/26 (水) 12:00-13:00 @ Livesense TechLunch
発表者:植村 健太
Livesense Inc.
PRO
April 21, 2014
More Decks by Livesense Inc.
See All by Livesense Inc.
Rubyはただの⾔語に⾮ず
livesense
PRO
0
560
28新卒_Webエンジニア職採用_会社説明資料
livesense
PRO
0
130
27新卒_総合職採用_会社説明資料
livesense
PRO
0
7.3k
27新卒_Webエンジニア職採用_会社説明資料
livesense
PRO
0
12k
株式会社リブセンス・転職会議 採用候補者様向け資料
livesense
PRO
0
590
株式会社リブセンス 会社説明資料(報道関係者様向け)
livesense
PRO
1
1.8k
データ基盤の負債解消のためのリプレイス
livesense
PRO
0
690
26新卒_総合職採用_会社説明資料
livesense
PRO
0
13k
株式会社リブセンス会社紹介資料 / Invent the next common.
livesense
PRO
2
75k
Other Decks in Technology
See All in Technology
行動するAIのためのオントロジー | DevRev — Encraft #26.pdf
dvrv_tknrszk
2
670
認知負荷を吸収し、プロダクトをまたぐPR Preview基盤の設計事例
taiki45
2
570
React Nativeでの OTA Updateって、 どう説明する?
ichiki1023
0
140
FinTech 1-2 : Overview of FinTech
ks91
PRO
0
130
The seven pitfalls of AI (revised version)
ufried
0
220
『GOエコノミー 』(相乗りサービス) におけるスペック駆動開発
mot_techtalk
1
220
Swap and Memory Reclaim - Squeezing Out More RAM
ennael
PRO
1
1.5k
形式手法を使って仕様をコーディングしよう
mikanichinose
0
170
AI駆動開発、viviONの1年 ── うまくいったこと・いかなかったこと
vivion
0
140
Codex概要
ymiya55
0
240
オブザーバビリティを高める AI エージェント体験を考える / Designing AI Agent Experiences That Enhance Observability
aoto
PRO
2
220
AI駆動開発で仕様はどこまで書くべきか? ― 人とAIの責務境界から考える開発プロセスの実践
takahiromatsui
1
300
Featured
See All Featured
Building Better People: How to give real-time feedback that sticks.
wjessup
370
20k
Applied NLP in the Age of Generative AI
inesmontani
PRO
4
2.5k
Faster Mobile Websites
deanohume
310
32k
Fantastic passwords and where to find them - at NoRuKo
philnash
52
3.9k
Done Done
chrislema
187
17k
Leveraging LLMs for student feedback in introductory data science courses - posit::conf(2025)
minecr
1
410
SEO in 2025: How to Prepare for the Future of Search
ipullrank
3
3.9k
Measuring & Analyzing Core Web Vitals
bluesmoon
9
1k
Dealing with People You Can't Stand - Big Design 2015
cassininazir
367
27k
The Director’s Chair: Orchestrating AI for Truly Effective Learning
tmiket
1
310
Deep Space Network (abreviated)
tonyrice
0
350
The SEO identity crisis: Don't let AI make you average
varn
0
570
Transcript
PQFO$7ͰإΛ͝ʹΐ͝ʹΐͯ͠ΈΔ σβΠϯ෦ɹ২ଜݐଠ 142݄26ਫ༵
PQFODWͱ OpenCVʢΦʔϓϯγʔϰΟɺӳޠ: Open Source Computer Vision LibraryʣͱΠϯς ϧ͕։ൃɾެ։ͨ͠ΦʔϓϯιʔεͷίϯϐϡʔλϏδϣϯ͚ϥΠϒϥϦɻ C/C++ɺJavaɺPython͚ʹϥΠϒϥϦ͕͋Δ ίϯϐϡʔλϏδϣϯʢcomputer
visionʣେࡶʹݴͬͯɺʮϩϘοτͷʯΛ࡞ΔݚڀͰ͋Δ 142݄26ਫ༵
Կ͕ग़དྷΔʁ ը૾ॲཧ (Image Processing) ߏղੳ (Structural Analysis) Ϟʔγϣϯղੳͱମ (Motion Analysis
and Object Tracking) ύλʔϯೝࣝ (Pattern Recognition) ΧϝϥΩϟϦϒϨʔγϣϯͱ3࣍ݩ࠶ߏ (Camera Calibration and 3D Reconstruction) ػցֶश ϢʔβΠϯλϑΣʔε 142݄26ਫ༵
ࠓը૾ೝࣝΛ͠·͢ɻ 142݄26ਫ༵
ը૾ೝࣝͬͯԿʁ ίϯϐϡʔλʔ͜ͷը૾إͳͷ͔ɺͳΜͳͷ͔Α͘Θ͔ͬͯͳ͍ͷ ͰͦΕΛڭ͑ͯ͋͛·͢ɻ 142݄26ਫ༵
Ͳ͏ͬͯೝ͍ࣝͯ͠Δʁ 1.ਓͷإͷࣸਅͬͯͷճΓΛ΅͔ͯ͠ݟΔͱɺͦͷ͙͢Լͷͷ͋ͨΓΑΓ҉͍ɻ 2.૭ը૾ͷத͔ΒͷลΓʹͳΔҐஔͷ্Լʹྡ͢ΔԣͷํܗྖҬΛΓग़ͯͦ͠ΕͧΕͷྖҬͷ ໌ͷฏۉΛͱΔɻ 3.্ͷྖҬ͕໌Δͯ͘ԼͷྖҬ͕҉͔ͬͨΒإը૾ͷީิʹ͢Δɻ 4.ඓےͷఱลΑΓඓےͷ྆ͷ໌Δ͞ͷฏۉͷ΄͏͕খ͍͜͞ͱ͕ଟ͍ɻ ʙ إͱͯ͠ೝࣝ ͲΜͲΜإͷީิ͔Β֎͍ͯ͘͠ 142݄26ਫ༵
࣮ࡍʹͬͯΈΔ XJUI 142݄26ਫ༵
PQFO'SBNFXPSLTͬ͘͟Γ ήʔϜͱ͔ΠϯελϨʔγϣϯΛ࡞Δͷʹ͍ͯΔ ΫϩεϓϥοτϑΥʔϜͷ ΞϓϦέʔγϣϯ։ൃϑϨʔϜϫʔΫ 142݄26ਫ༵
142݄26ਫ༵
142݄26ਫ༵
إݕग़ϥΠϒϥϦΛ͏ PG0QFO$7 142݄26ਫ༵
PGY0QFO$Wͷإݕग़ϥΠϒϥϦ ϝϞɿϓϩάϥϜಈ͔͢ 142݄26ਫ༵
void testApp::setup(){ finder.setup("haarcascade_frontalface_alt2.xml"); vid.allocate(CAM_WIDTH, CAM_HEIGHT); gray.allocate(CAM_WIDTH, CAM_HEIGHT); vidGrabber.setVerbose(true); vidGrabber.initGrabber(CAM_WIDTH, CAM_HEIGHT);
} PQFO$7ͷإݕग़ϥΠϒϥϦ 142݄26ਫ༵
void testApp::update(){ vidGrabber.update(); vid.setFromPixels(vidGrabber.getPixels(), CAM_WIDTH, CAM_HEIGHT); vid.convertRgbToHsv(); vid.convertToGrayscalePlanarImage(gray, 2); finder.findHaarObjects(gray);
} 142݄26ਫ༵
void testApp::draw(){ ofSetColor(255,255,255); vidGrabber.draw(0,0); gray.draw(CAM_WIDTH, 0, CAM_WIDTH, CAM_HEIGHT); //ofNoFill(); ofSetColor(255,
255, 0); int numFace = finder.blobs.size(); float x[numFace], y[numFace], w[numFace], h[numFace]; for(int i = 0; i < numFace; i++) { ofRect( finder.blobs[i].boundingRect ); } } 142݄26ਫ༵
ྠֲநग़Λߦ͏ʴিಥఆ PG#PYE PG0QFO$7 142݄26ਫ༵
എܠࠩΛߦ͏ࣄͰྠֲநग़ 142݄26ਫ༵
ͬ͘͟ΓͲ͏͍͏͜ͱ͔ ݕग़͢Δମ͕ͳ͍ঢ়ଶͰɺഎܠը૾Λه͓ͯ͘͠ ݕग़͢ΔମΛؚΊͨݱࡏͷը૾Λೖྗ ه͓͍ͯͨ͠എܠը૾ͱɺݱࡏͱͷࠩΛͱͬͨը૾Λੜ ࠩΛͱͬͨը૾Λഎܠ͔Β͢ΔͨΊɺઃఆͨ͠Λᮢͱ͠ ͯɺനࠇʹۃԽ͢Δ 142݄26ਫ༵
நग़ͨ͠ϥΠϯΛিಥఆʹར༻͢Δ 142݄26ਫ༵
PGY#PYE 142݄26ਫ༵
PGY#PYEͬ͘͟Γ ʮཧΤϯδϯʯ ॏྗিಥɺຎࡲͱ͍ͬͨཧܭࢉΛෳࡶͳܭࢉΛ͢Δ͜ͱͳ͘ར༻Ͱ͖Δ //Box2Dॳظઃఆ box2d.init(); //Box2DͷੈքΛॳظԽ box2d.setGravity(xํͷॏྗ, yํͷॏྗ); box2d.createBounds(0, 0,ɹը໘ͷ෯,ɹը໘ͷߴ͞);
//ը໘ΛนͰғΉ box2d.setFPS(30); //30fpsͰදࣔ box2d.checkBounds(true); Person yuji; yuji.setPhysics(ॏ͞,ൃྗ,ຎࡲྗ); yuji.setup(box2d.getWorld(), x࠲ඪ, y࠲ඪ (, r)); yuji.draw(); #include "ofxBox2d.h" ofxBox2d box2d; vector <ofxBox2dCircle> Person;ɹ//ಈతྻΛ༻͢ΔɻϜμͳϝϞϦফඅΛ͙ ઃఆྫϔομϑΝΠϧଆ ઃఆྫιʔεϑΝΠϧଆ ϝϞɿϓϩάϥϜಈ͔͢ 142݄26ਫ༵
·ͱΊ ofxOpenCvݕग़ϥΠϒϥϦxmlΛಡΈࠐΉ͚ͩͳͷͰݪཧͱ͔͔Βͳͯ͑͘Δɻ͍Ζ͍Ζ Έ߹Θͤͳ͍ͱਫ਼ଟ͕͋Μͳ͍ إೝূϋʔυϧߴ͍ͬ͢ɻຊՈͰͬͨ΄͏͕Αͦ͞͏ addon͕๛ͰɺΈ߹Θͤ࣍ୈͰ͍Ζ͍ΖͰ͖ΔʢMIDI৴߸+ཧԋࢉ+KINECTͱ͔ʣ Έ߹Θͤͳ͍ͱHTML5ͷAPIͱ͋Μ·มΘΜͳ͍ɻ ϝϞϦϦʔΫͨ͠ΒམͪΔɻݶք͕ͲΕ͘Β͍ͳͷ͔Ϗϧυͯ͠Έͳ͍ͱ͔Βͳ͍͠ɺϚγ ϯεϖοΫʹґଘ ͡Ί͔Β៛ີʹઃܭ࣮ͯ͢͠ΔΑΓɺ࡞Γͳ͕Βࢼ͢ʢεέον͢Δʣײ͡Ͱઓͬͨ΄͏ ͕͍͍ɻܭࢉͲ͏Γʹ͍͔ͳ͍͜ͱͷ΄͏͕ଟ͍ɻ
͋·ΓݫີͳϓϩάϥϜؾʹ͠ͳ͍͍ͯ͘Α͏ʹྑ͖ʹܭΒͬͯ͘ΕͯΔͷͰɺϓϩάϥϜ͔ Μͳͯ͘ͳΜͱ͔ͳΔɻ 142݄26ਫ༵
͓ΘΓ openframeworksͰΩϛिϓϩάϥϛϯάʂʂ 142݄26ਫ༵