Upgrade to Pro — share decks privately, control downloads, hide ads and more …

Intro to ConvNets: The backbone of modern compu...

Avatar for Tryolabs Tryolabs
April 10, 2019
Technology
2.3k
0

Intro to ConvNets: The backbone of modern computer vision

Avatar for Tryolabs

Tryolabs

April 10, 2019

More Decks by Tryolabs

See All by Tryolabs
An Introduction to Machine Learning and How to Teach Machines to See
Avatar for Tryolabs tryolabs
0
2.3k
Tryolabs Workshop: Object Detection with Deep Learning
Avatar for Tryolabs tryolabs
0
510
PyImageConf Workshop: Object Detection with Deep Learning
Avatar for Tryolabs tryolabs
0
300
Introduction to Object Detection - PyCon APAC 2018
Avatar for Tryolabs tryolabs
0
2.7k
Building an Object Detection toolkit with TensorFlow (ODSC West 2017)
Avatar for Tryolabs tryolabs
1
2.6k
Building an Object Detection toolkit with TensorFlow (PyLadies Meetup)
Avatar for Tryolabs tryolabs
1
420
Building an Object Detection toolkit with TensorFlow (ODSC Europe 2017)
Avatar for Tryolabs tryolabs
1
220
Tryolabs Working Trip NYC 2017 in pictures.
Avatar for Tryolabs tryolabs
0
1.7k
Machine Learning 101 - Tryolabs
Avatar for Tryolabs tryolabs
0
230

Other Decks in Technology

See All in Technology
AIに障害切り分けを全部やってもらった。 。 。 。
Avatar for estie | エスティ estie
0
300
40代で“やっとエンジニアになれた”――閉じた学びを開き、空の青さを知る / 20260628 Naoki Takahashi
Avatar for SHIFT EVOLVE shift_evolve
PRO
4
1.4k
Kotlin 開発のツラミを爆破した話！ / Explode the difficulty of Kotlin dev!
Avatar for Kengo TODA eller86
0
120
デジタル・デザイン構想 by Sayaka Ishizuka
Avatar for Sayaka Ishizuka y150saya
0
150
Amazon Redshift zero-ETL 統合を活用した軽量なマルチプロダクトデータ可視化基盤 / Lightweight Multi-Product Data Visualization with Amazon Redshift Zero-ETL
Avatar for 株式会社カミナシ kaminashi
0
140
時期が悪い！それでもRaspberry Piを買って遊んで活用するには / 20260627-osc26do-rpi-jikigawarui
Avatar for Akira Ouchi akkiesoft
1
950
MySQL & MySQL HeatWave Report - June 2026
Avatar for Olivier DASINI freshdaz
0
250
そこにあるから地図ができる～位置を示す"モノ"を愉しむ～ - Interface 2026年6月号GPS特集オフ会 / interface_202606_GPS_offline
Avatar for sakaik sakaik
1
160
Multi-Agent並列開発を 安全に回すための技術 / Technology for Safely Multi-Agent Parallel Development
Avatar for philomagi tooppoo
0
230
起点・思考・出力で分解する 〜PM業務の自動化設計〜
Avatar for kazu_kichi_67 kazu_kichi_67
2
1.2k
AIをフル活用してオンコール機能のプロトタイプを2日で作った話 / Building an AI-Powered On-Call Prototype in Just Two Days
Avatar for Narimichi Takamura nari_ex
0
160
サイバーエージェントにおけるAI推進戦略と変革への取り組み
Avatar for shotaTsuge shotatsuge
0
640

Featured

See All Featured
Lessons Learnt from Crawling 1000+ Websites
Avatar for Charles Meaden charlesmeaden
PRO
1
1.3k
Ethics towards AI in product and experience design
Avatar for Skipper Chong Warson skipperchong
2
320
My Coaching Mixtape
Avatar for mlcsv mlcsv
0
160
How Software Deployment tools have changed in the past 20 years
Avatar for Geshan Manandhar geshan
0
34k
[Rails World 2023 - Day 1 Closing Keynote] - The Magic of Rails
Avatar for Eileen M. Uchitelle eileencodes
38
2.9k
Facilitating Awesome Meetings
Avatar for Lara Hogan lara
57
7k
Max Prin - Stacking Signals: How International SEO Comes Together (And Falls Apart)
Avatar for Tech SEO Connect techseoconnect
PRO
0
190
Why Mistakes Are the Best Teachers: Turning Failure into a Pathway for Growth
Avatar for Umar Saidu Auna auna
0
170
Unlocking the hidden potential of vector embeddings in international SEO
Avatar for Frank van Dijk frankvandijk
0
860
Digital Projects Gone Horribly Wrong (And the UX Pros Who Still Save the Day) - Dean Schuster
Avatar for UX Y'all uxyall
1
1.8k
Organizational Design Perspectives: An Ontology of Organizational Design Elements
Avatar for Dr. Kim W Petersen kimpetersen
PRO
1
760
The SEO identity crisis: Don't let AI make you average
Avatar for Varn varn
0
510

Transcript

  1. The backbone of modern computer vision Intro to ConvNets

  2. 2

  3. 3 =

  4. 4

  5. Convolutional Network

  6. 6 ⊙ = Σ input output kernel

  7. 7 input output ⊙ = kernel Σ

  8. 8 input output = kernel Σ ⊙

  9. 9 input output ⊙ = kernel Σ

  10. 10 kernel input output

  11. 11

  12. 12 Optimization Label: Bird ConvNet Loss Function Prediction: Cat

  13. 13 Optimization ConvNet Prediction: Bird Loss Function Label: Bird

  14. 14 Non-linear function from: wikipedia.org Sigmoid Function

  15. 15 Pooling operation from: computersciencewiki.org

  16. 16

  17. 17 Conv layer Non-linear function Convolution & Non-linear function &

    Pooling Pooling Conv layer Non-linear function Pooling Conv layer Non-linear function

  18. 18 AlexNet (2012)

  19. Thanks!