Upgrade to Pro
— share decks privately, control downloads, hide ads and more …
Speaker Deck
Features
Speaker Deck
PRO
Sign in
Sign up for free
Search
Search
Intro to Computer Vision (for 4th Graders)
Search
Sam Liu
April 29, 2013
Education
2
290
Intro to Computer Vision (for 4th Graders)
Part of an outreach initiative by the Moon Goons microgravity research team at UIUC
Sam Liu
April 29, 2013
Tweet
Share
Other Decks in Education
See All in Education
AI for Learning
fonylew
0
180
Info Session MSc Computer Science & MSc Applied Informatics
signer
PRO
0
210
OJTに夢を見すぎていませんか? ロールプレイ研修の試行錯誤/tryanderror-in-roleplaying-training
takipone
1
220
20250807_がんばらないコミュニティ運営
ponponmikankan
0
160
大学院進学について(2025年度版)
imash
0
110
20250625_なんでもCopilot 一年の振り返り
ponponmikankan
0
350
AIの時代こそ、考える知的学習術
yum3
2
190
”育てる”から”育つ”仕組みへ!スクラムによる新入社員教育
arapon
0
130
~キャラ付け考えていますか?~ AI時代だからこそ技術者に求められるセルフブランディングのすゝめ
masakiokuda
7
410
教える側は、初学者に谷越えまで伴走すべき(ダニング・クルーガー効果からの考察)
hysmrk
3
130
『会社を知ってもらう』から『安心して活躍してもらう』までの プロセスとフロー
sasakendayo
0
260
Course Review - Lecture 12 - Next Generation User Interfaces (4018166FNR)
signer
PRO
0
1.8k
Featured
See All Featured
The Cult of Friendly URLs
andyhume
79
6.6k
The MySQL Ecosystem @ GitHub 2015
samlambert
251
13k
Unsuck your backbone
ammeep
671
58k
Building an army of robots
kneath
306
46k
Docker and Python
trallard
45
3.5k
Become a Pro
speakerdeck
PRO
29
5.5k
Rebuilding a faster, lazier Slack
samanthasiow
83
9.1k
Typedesign – Prime Four
hannesfritz
42
2.8k
Raft: Consensus for Rubyists
vanstee
140
7.1k
A Modern Web Designer's Workflow
chriscoyier
696
190k
Large-scale JavaScript Application Architecture
addyosmani
512
110k
StorybookのUI Testing Handbookを読んだ
zakiyama
30
6.1k
Transcript
Making a computer see
Who’s this?
None
None
Each box has 1 color YOU see a color! COMPUTER
sees a number! What are the primary colors?
None
None
Tells us how much red, green, and blue is in
the pixel. MAXIMUM = 255 MINIMUM = 0 “Mix” the lights!
None
This is cool. Why? We can make cool things! Like
what? A rainbow clock! http://ambushnetworks.com/clock/
How do I make a computer find something?
http://harthur.github.io/kittydar/ (2) Steps “Training” “Detection”
Take lots of pictures of cats Take lots
of pictures of not cats
Calculate a special thing called “histogram of oriented gradients”
(it’s hard math)
Give data to “neural network” (more hard math)
Run “backpropogation” (so much hard math) Now your neural network can detect cats!
Slide a window over the image Keeping asking
neural network “hey does this little picture look like a cat?” A little slow.
Is easier than detecting cats! Why? Oil on
our faces and symmetry (short answer)
Single image face detection http://liuliu.me/ccv/js/nss/ Silly glasses via
face detection http://youtu.be/YYES9Qd094o?t=1m20s Eulerian Video Magnification http://youtu.be/ONZcjs1Pjmk?t=1m35s
Our quadrocopter has a camera! I teach it to see
things and to know “I need to spin” or “I need to stop spinning” It’s a cool job! http://youtu.be/TbBM1Gpt390?t=9s
If you want to have fun making stuff like
this when you get older, STUDY MATH. MATH IS FUN. It doesn’t seem that way sometimes, but it’s totally worth it in the end! Computer vision can make fun games! Computer vision can help people!
Questions?