畳み込みニューラルネットワークを用いた自動車の三次元モデルにおける各車型の特徴抽出と視覚化

Translated title of the contribution: Analyses and Visualization of Characteristics of Car Body Types by Using Convolutional Neural Network

田中 俊太朗, 原田 利宣, 小野 謙二

Research output: Contribution to journalArticle

Abstract

The cars are classified by cars' body types. However the characteristics are basically similar at first sight, so it is difficult to distinguish the differences among those cars' body types. Therefore, in this study, we considered that cars' characteristics could be analyzed by using deep learning and image recognition technology, developed a system to visualize the judgment and characteristic parts of cars' body types. Specifically, we made renderings of the CG model of 30 cars by setting 360 viewpoints in 1 degree increments around each car. Deep learning was performed using these 2D images as teacher signals. The car body type recognition probability of each angle is graphed, and the characteristic parts of each car body type are visualized. As a result, we clarified the visual angles and the pars contributing the judgment of cars' body types.
Original languageJapanese
Pages (from-to)113-121
Number of pages9
Journal日本感性工学会論文誌
Volume18
Issue number1
DOIs
Publication statusPublished - 2019

Fingerprint

Railroad cars
Visualization
Neural networks
Image recognition

Cite this

畳み込みニューラルネットワークを用いた自動車の三次元モデルにおける各車型の特徴抽出と視覚化. / 田中俊太朗; 原田利宣; 小野謙二.

In: 日本感性工学会論文誌, Vol. 18, No. 1, 2019, p. 113-121.

Research output: Contribution to journalArticle

@article{7bfa5b08a4534359b9cd0402b282dfdc,
title = "畳み込みニューラルネットワークを用いた自動車の三次元モデルにおける各車型の特徴抽出と視覚化",
abstract = "The cars are classified by cars' body types. However the characteristics are basically similar at first sight, so it is difficult to distinguish the differences among those cars' body types. Therefore, in this study, we considered that cars' characteristics could be analyzed by using deep learning and image recognition technology, developed a system to visualize the judgment and characteristic parts of cars' body types. Specifically, we made renderings of the CG model of 30 cars by setting 360 viewpoints in 1 degree increments around each car. Deep learning was performed using these 2D images as teacher signals. The car body type recognition probability of each angle is graphed, and the characteristic parts of each car body type are visualized. As a result, we clarified the visual angles and the pars contributing the judgment of cars' body types.",
author = "俊太朗 田中 and 利宣 原田 and 謙二 小野",
year = "2019",
doi = "10.5057/jjske.TJSKE-D-18-00039",
language = "Japanese",
volume = "18",
pages = "113--121",
journal = "日本感性工学会論文誌",
issn = "1882-8930",
publisher = "日本感性工学会",
number = "1",

}

TY - JOUR

T1 - 畳み込みニューラルネットワークを用いた自動車の三次元モデルにおける各車型の特徴抽出と視覚化

AU - 田中, 俊太朗

AU - 原田, 利宣

AU - 小野, 謙二

PY - 2019

Y1 - 2019

N2 - The cars are classified by cars' body types. However the characteristics are basically similar at first sight, so it is difficult to distinguish the differences among those cars' body types. Therefore, in this study, we considered that cars' characteristics could be analyzed by using deep learning and image recognition technology, developed a system to visualize the judgment and characteristic parts of cars' body types. Specifically, we made renderings of the CG model of 30 cars by setting 360 viewpoints in 1 degree increments around each car. Deep learning was performed using these 2D images as teacher signals. The car body type recognition probability of each angle is graphed, and the characteristic parts of each car body type are visualized. As a result, we clarified the visual angles and the pars contributing the judgment of cars' body types.

AB - The cars are classified by cars' body types. However the characteristics are basically similar at first sight, so it is difficult to distinguish the differences among those cars' body types. Therefore, in this study, we considered that cars' characteristics could be analyzed by using deep learning and image recognition technology, developed a system to visualize the judgment and characteristic parts of cars' body types. Specifically, we made renderings of the CG model of 30 cars by setting 360 viewpoints in 1 degree increments around each car. Deep learning was performed using these 2D images as teacher signals. The car body type recognition probability of each angle is graphed, and the characteristic parts of each car body type are visualized. As a result, we clarified the visual angles and the pars contributing the judgment of cars' body types.

U2 - 10.5057/jjske.TJSKE-D-18-00039

DO - 10.5057/jjske.TJSKE-D-18-00039

M3 - 記事

VL - 18

SP - 113

EP - 121

JO - 日本感性工学会論文誌

JF - 日本感性工学会論文誌

SN - 1882-8930

IS - 1

ER -