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
一般物体検出とLSTMを用いた画像に基づく屋内位置推定 - IPSJ UBI82
Search
Aokiti
May 12, 2024
0
450
一般物体検出とLSTMを用いた画像に基づく屋内位置推定 - IPSJ UBI82
http://id.nii.ac.jp/1001/00233750/
Aokiti
May 12, 2024
Tweet
Share
More Decks by Aokiti
See All by Aokiti
[d-hacks Docker講座] Dockerで動かすローカルLLM入門
sakusaku3939
0
43
[論文輪読会] A survey of model compression strategies for object detection
sakusaku3939
0
20
[論文輪読会] ViT-1.58b
sakusaku3939
0
150
d-hacks PyTorchモデル実装会 2024f
sakusaku3939
0
47
[論文輪読会] Binarized Neural Networks
sakusaku3939
0
57
MoodTune 東京AI祭ハッカソン決勝
sakusaku3939
0
590
d-hacks PyTorch実装会 2023f
sakusaku3939
0
30
[DL勉強会] 第5章 ディープラーニングを活用したアプリケーション 後半
sakusaku3939
0
17
Presc - 高校研究発表会
sakusaku3939
0
1.7k
Featured
See All Featured
Building a Modern Day E-commerce SEO Strategy
aleyda
45
8.7k
Building Flexible Design Systems
yeseniaperezcruz
330
40k
Gemini Prompt Engineering: Practical Techniques for Tangible AI Outcomes
mfonobong
2
280
How To Stay Up To Date on Web Technology
chriscoyier
791
250k
Code Review Best Practice
trishagee
74
20k
Marketing to machines
jonoalderson
1
4.7k
The B2B funnel & how to create a winning content strategy
katarinadahlin
PRO
1
280
For a Future-Friendly Web
brad_frost
182
10k
Code Reviewing Like a Champion
maltzj
527
40k
世界の人気アプリ100個を分析して見えたペイウォール設計の心得
akihiro_kokubo
PRO
66
37k
Digital Projects Gone Horribly Wrong (And the UX Pros Who Still Save the Day) - Dean Schuster
uxyall
0
400
SEO Brein meetup: CTRL+C is not how to scale international SEO
lindahogenes
0
2.4k
Transcript
1
2
3
4
5
6
[7] F. Walch, C. Hazirbas, L. Leal-Taixé, T. Sattler, S.
Hilsenbeck, D. Cremers 7 https://doi.org/10.1109/ICCV.2017.75
8 [8] S Nilwong, D Hossain, S Kaneko, G Capi
https://doi.org/10.3390/machines7020025
9
10 CNN(GoogLeNetモデル) LSTM(次元を削減) CNN CNN CNN LSTM(1つの特徴量に変換)
11 CNN CNN CNN LSTM(1つの特徴量に変換)
12
13
14
15
16 𝑀𝐴𝐸 = 1 𝑛 𝑖=1 𝑛 |ෝ 𝑥𝑖
− 𝑥𝑖 + | ෝ 𝑦𝑖 − 𝑦𝑖 |)
17 CNN (4層) CNN(4層) CNN(4層) LSTM(1つの特徴量に変換) CNN(4層) CNN(GoogLeNet) CNN(GoogLeNet) LSTM(次元を削減)
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34