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
40
[論文輪読会] A survey of model compression strategies for object detection
sakusaku3939
0
12
[論文輪読会] ViT-1.58b
sakusaku3939
0
140
d-hacks PyTorchモデル実装会 2024f
sakusaku3939
0
47
[論文輪読会] Binarized Neural Networks
sakusaku3939
0
48
MoodTune 東京AI祭ハッカソン決勝
sakusaku3939
0
580
d-hacks PyTorch実装会 2023f
sakusaku3939
0
29
[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
Evolution of real-time – Irina Nazarova, EuRuKo, 2024
irinanazarova
9
1.1k
Raft: Consensus for Rubyists
vanstee
141
7.3k
What Being in a Rock Band Can Teach Us About Real World SEO
427marketing
0
160
SEO for Brand Visibility & Recognition
aleyda
0
4.1k
コードの90%をAIが書く世界で何が待っているのか / What awaits us in a world where 90% of the code is written by AI
rkaga
58
41k
Unlocking the hidden potential of vector embeddings in international SEO
frankvandijk
0
140
RailsConf 2023
tenderlove
30
1.3k
Leveraging LLMs for student feedback in introductory data science courses - posit::conf(2025)
minecr
0
100
The Director’s Chair: Orchestrating AI for Truly Effective Learning
tmiket
1
74
実際に使うSQLの書き方 徹底解説 / pgcon21j-tutorial
soudai
PRO
196
71k
Measuring & Analyzing Core Web Vitals
bluesmoon
9
720
Docker and Python
trallard
47
3.7k
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