國立虎尾科技大學 |

Scalable Human Identification with Deep Learning.

紀錄類型:	書目-語言資料,手稿 : Monograph/item
正題名/作者:	Scalable Human Identification with Deep Learning./
作者:	Xiao, Tong.
面頁冊數:	1 online resource (99 pages)
附註:	Source: Dissertation Abstracts International, Volume: 79-07(E), Section: B.
標題:	Computer science. -
電子資源:	click for full text (PQDT)
ISBN:	9780355554151

Scalable Human Identification with Deep Learning.
Xiao, Tong.

Scalable Human Identification with Deep Learning. - 1 online resource (99 pages)

Source: Dissertation Abstracts International, Volume: 79-07(E), Section: B.

Thesis (Ph.D.)--The Chinese University of Hong Kong (Hong Kong), 2017.

Includes bibliographical references

Human identification, which aims at finding a target person of interest from a gallery of digital photos, is one of the fundamental problems in computer vision. It is a key component in many real-world applications including smart phone apps, self-driving cars, home security systems, and intelligent surveillance cameras.

Electronic reproduction.
Ann Arbor, Mich. :
ProQuest,
2018

Mode of access: World Wide Web

ISBN: 9780355554151Subjects--Topical Terms:

573171
Computer science.
Index Terms--Genre/Form:

554714
Electronic books.

Scalable Human Identification with Deep Learning.
LDR:03028ntm a2200337K 4500 001 912204
005 20180608102941.5
006 m o u
007 cr mn||||a|a||
008 190606s2017 xx obm 000 0 eng d
020 $a 9780355554151
035 $a (MiAaPQ)AAI10753325
035 $a AAI10753325
040 $a MiAaPQ $b eng $c MiAaPQ
100 1 $a Xiao, Tong. $3 1184455
245 1 0 $a Scalable Human Identification with Deep Learning.
264 0 $c 2017
300 $a 1 online resource (99 pages)
336 $a text $b txt $2 rdacontent
337 $a computer $b c $2 rdamedia
338 $a online resource $b cr $2 rdacarrier
500 $a Source: Dissertation Abstracts International, Volume: 79-07(E), Section: B.
500 $a Adviser: Xiaogang Wang.
502 $a Thesis (Ph.D.)--The Chinese University of Hong Kong (Hong Kong), 2017.
504 $a Includes bibliographical references
520 $a Human identification, which aims at finding a target person of interest from a gallery of digital photos, is one of the fundamental problems in computer vision. It is a key component in many real-world applications including smart phone apps, self-driving cars, home security systems, and intelligent surveillance cameras.
520 $a Thanks to the development of deep learning research and large-scale well annotated datasets, deep neural networks are now capable of recognizing thousands of object categories. However, human identification is still challenging because it lacks a dataset large enough to supervise the model training. Moreover, existing small datasets usually have their own image biases, which makes it hard to learn a single model that generalizes over all these domains. Meanwhile, most of the existing research simplified the problem setting, which leaves a gap between research approaches and practical applications.
520 $a In this dissertation we address these challenges from three aspects to make human identification scalable to real-world data and applications. First, we propose a semisupervised deep learning framework that uses noisy-labeled rather than well annotated data. We collect a large-scale clothing dataset with noisy annotations, from which we can learn good representations for clothes that help recognize human. Second, we develop a joint single task learning algorithm and a domain guided dropout technique to learn a single model from multiple human identification datasets with domain biases. It enables us to collectively use the data contributed by different people in the community. At last, we focus on the more realistic problem setting that finds a target person in whole scene images. We develop a unified framework that combines person detection and identification, as well as a loss function that trains the identification model effectively.
533 $a Electronic reproduction. $b Ann Arbor, Mich. : $c ProQuest, $d 2018
538 $a Mode of access: World Wide Web
650 4 $a Computer science. $3 573171
650 4 $a Artificial intelligence. $3 559380
655 7 $a Electronic books. $2 local $3 554714
690 $a 0984
690 $a 0800
710 2 $a ProQuest Information and Learning Co. $3 1178819
710 2 $a The Chinese University of Hong Kong (Hong Kong). $b Electronic Engineering. $3 1184454
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=10753325 $z click for full text (PQDT)