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
三次元形状とディープラーニング
Search
Tatsuya Yatagawa
December 21, 2020
Technology
0
270
三次元形状とディープラーニング
三次元形状処理とディープラーニングの初歩についてまとめたスライドです。
Tatsuya Yatagawa
December 21, 2020
Tweet
Share
More Decks by Tatsuya Yatagawa
See All by Tatsuya Yatagawa
Event-based Camera Simulation using Monte Carlo Path Tracing with Adaptive Denoising (ICIP 2023)
tatsy
0
480
VC'23 チュートリアル講演 『三次元メッシュと深層学習』
tatsy
0
190
SIGGRAPH 2022: Neural Geometry Processing
tatsy
0
210
Real-time Shading with Free-form Planar Area Lights using Linearly Transformed Cosines (JCGT & I3D 2022)
tatsy
0
910
SIGGRAPH Asia 2021: Real-Time Rendering
tatsy
0
88
SIGGRAPH 2021: Denoising and Guiding
tatsy
0
62
SIGGRAPH 2021: Shape Generation
tatsy
0
150
SIGGRAPH 2021: Meshing
tatsy
0
91
12. Keyboard & Mouse
tatsy
0
190
Other Decks in Technology
See All in Technology
[CVPR2025論文読み会] Linguistics-aware Masked Image Modelingfor Self-supervised Scene Text Recognition
s_aiueo32
0
210
JOAI発表資料 @ 関東kaggler会
joai_committee
1
260
OpenAPIから画面生成に挑戦した話
koinunopochi
0
150
TypeScript入門
recruitengineers
PRO
8
1.5k
モダンフロントエンド 開発研修
recruitengineers
PRO
2
260
キャリアを支え組織力を高める「多層型ふりかえり」 / 20250821 Kazuki Mori
shift_evolve
PRO
2
300
AIドリブンのソフトウェア開発 - うまいやり方とまずいやり方
okdt
PRO
9
570
Figma + Storybook + PlaywrightのMCPを使ったフロントエンド開発
yug1224
4
210
Goでマークダウンの独自記法を実装する
lag129
0
210
Product Management Conference -AI時代に進化するPdM-
kojima111
0
220
株式会社ARAV 採用案内
maqui
0
340
自社製CMSからmicroCMSへのリプレースがプロダクトグロースを加速させた話
nextbeatdev
0
130
Featured
See All Featured
JavaScript: Past, Present, and Future - NDC Porto 2020
reverentgeek
50
5.5k
Understanding Cognitive Biases in Performance Measurement
bluesmoon
29
1.8k
Optimizing for Happiness
mojombo
379
70k
Music & Morning Musume
bryan
46
6.7k
Helping Users Find Their Own Way: Creating Modern Search Experiences
danielanewman
29
2.8k
The Pragmatic Product Professional
lauravandoore
36
6.8k
Designing Experiences People Love
moore
142
24k
"I'm Feeling Lucky" - Building Great Search Experiences for Today's Users (#IAC19)
danielanewman
229
22k
The Cult of Friendly URLs
andyhume
79
6.5k
Creating an realtime collaboration tool: Agile Flush - .NET Oxford
marcduiker
31
2.2k
Building Flexible Design Systems
yeseniaperezcruz
328
39k
StorybookのUI Testing Handbookを読んだ
zakiyama
30
6k
Transcript
None
• 機械学習 ニューラルネット 深層学習
• • • θ • = ; = 2 2
+ 1 + 0 ෩ = arg min =0 −1 ( , ; )
• • • 0 = 0 + 0 1 =
1 + 1 2 = 2 + 2 = 0 0 + 1 1 + 2 2 + = 0 0 + 1 1 + 2 2 +(0 0 + 1 1 + 2 2 + )
• • • 0 1 2 0 = ReLU(0 +
0 ) 1 = ReLU(1 + 1 ) 2 = ReLU(2 + 2 )
= 32 = 128 = 512 • •
• • = 512 = 1 = = 4
•
•
•
•
•
•
• • • • • • • • • •
•
(c) Velodyne (c) Andrew Tallon (c) Microsoft (c) Sony
• , , = 0
′ = − ′ = + ′
• ☺ ☺ ☺ ☺ ,
, = 0
☺ ☺ ☺ ☺ , ,
= 0
• • • • •
• • • •
•
•
• ⚫ ⚫ ⚫
• • •
• • •
• •
• • •
• • • • • • •
☺ ☺ ☺ ☺ , ,
= 0
• • • (c) Velodyne
• • • (0 , 0 , 0 ) (1
, 1 , 1 ) ( , , ) (−1 , −1 , −1 ) (+1 , +1 , +1 ) (0 , 0 , 0 ) (1 , 1 , 1 ) (−1 , −1 , −1 ) (+1 , +1 , +1 ) ( , , )
• • •
None
⚫
⚫
• •
• • •
•
•
•
• • • • • • •
☺ ☺ ☺ ☺ , ,
= 0
• • •
• • • •
• • ∗ = ℱ−1(ℱ ⋅ ℱ ) ′ =
ℱ−1(Θ ∘ ℱ )
• = 0 1 1 0 1 0 1 0
1 1 0 1 0 0 1 0 = − − 1 2− 1 2 = :
• • • = T
• () ′ = ℱ−1(Θ ∘ ℱ ) ′ =
ℱ = () ℱ−1 = ()
• • • +1 +2 +1 = ReLU ⋅
• • •
• • •
• •
• • •
• • •
• •
• • • • • •
☺ ☺ ☺ ☺ , ,
= 0
• , , =
• • •
• • • •
• •
• • •
• • •
• • • •
•
• • • •
• • •
• • • •
• • •
• • • •
• • • • • • • •
• • • • • • • • •