國立虎尾科技大學 |

Computer Vision Methods for Enhancing Remote Sighted Assistance.

紀錄類型:	書目-語言資料,手稿 : Monograph/item
正題名/作者:	Computer Vision Methods for Enhancing Remote Sighted Assistance./
作者:	Yu, Rui.
面頁冊數:	1 online resource (177 pages)
附註:	Source: Dissertations Abstracts International, Volume: 85-05, Section: B.
Contained By:	Dissertations Abstracts International85-05B.
標題:	Information science. -
電子資源:	click for full text (PQDT)
ISBN:	9798380730495

Computer Vision Methods for Enhancing Remote Sighted Assistance.
Yu, Rui.

Computer Vision Methods for Enhancing Remote Sighted Assistance. - 1 online resource (177 pages)

Source: Dissertations Abstracts International, Volume: 85-05, Section: B.

Thesis (Ph.D.)--The Pennsylvania State University, 2023.

Includes bibliographical references

Remote sighted assistance (RSA) has emerged as a conversational assistive technology for people with visual impairments (PVI), where a remote sighted agent interprets the real-time video feed coming from a visually impaired user's smartphone camera while conversing with the user to provide assistance. Due to its significant benefits for PVI, the RSA technology has attracted increasing attention from both academia and industry.With the growing complexity of tasks, researchers have found that reliance on smartphones' camera feed can be a limiting factor and reported several challenges for the agents, e.g., unfamiliarity with PVI's physical surroundings, difficulty in tracking moving objects, and interpreting various information in real-time. Since these challenges are related to understanding 3D environments and semantics from the live video, they can be potentially addressed by computer vision (CV) technologies. This dissertation aims to develop CV methods to enhance RSA services. The study has been carried out in three phases.In the first phase, we conducted a literature review and interviewed twelve RSA users to thoroughly understand the challenges in RSA. We summarized a comprehensive list of challenges, analyzed the causes and the technological needs, and thus identified some major CV problems in RSA. The second phase of the study is developing new methods to address key issues in the identified CV problems. In particular, we focus on addressing four representative CV problems: predicting the trajectories of pedestrians, recognizing persons or objects from camera feed, expanding the field of view of camera feed, and estimating the scale and depth in dual camera feed. The third phase is developing user-friendly RSA prototypes and verifying the CV methods through user studies. In our study, we focus on the exemplar problem of 3D indoor mapping and real-time localization. Specifically, we created an RSA prototype with ARKit-based iOS app for the third-phase study. Localizing RSA users on 3D maps in real-time has been proved effective for enhancing RSA services in our study. Through the three-phase study, we identified the challenges and CV problems in RSA, developed new methods for representative CV problems, and verified the effectiveness of exemplar CV techniques for supporting RSA.

Electronic reproduction.
Ann Arbor, Mich. :
ProQuest,
2024

Mode of access: World Wide Web

ISBN: 9798380730495Subjects--Topical Terms:

561178
Information science.
Subjects--Index Terms:

Remote sighted assistanceIndex Terms--Genre/Form:

554714
Electronic books.

Computer Vision Methods for Enhancing Remote Sighted Assistance.
LDR:03697ntm a22003857 4500 001 1148334
005 20240924101856.5
006 m o d
007 cr bn ---uuuuu
008 250605s2023 xx obm 000 0 eng d
020 $a 9798380730495
035 $a (MiAaPQ)AAI30720551
035 $a (MiAaPQ)PennState_23128rzy54
035 $a AAI30720551
040 $a MiAaPQ $b eng $c MiAaPQ $d NTU
100 1 $a Yu, Rui. $3 1474279
245 1 0 $a Computer Vision Methods for Enhancing Remote Sighted Assistance.
264 0 $c 2023
300 $a 1 online resource (177 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: Dissertations Abstracts International, Volume: 85-05, Section: B.
500 $a Advisor: Huang, Sharon X.
502 $a Thesis (Ph.D.)--The Pennsylvania State University, 2023.
504 $a Includes bibliographical references
520 $a Remote sighted assistance (RSA) has emerged as a conversational assistive technology for people with visual impairments (PVI), where a remote sighted agent interprets the real-time video feed coming from a visually impaired user's smartphone camera while conversing with the user to provide assistance. Due to its significant benefits for PVI, the RSA technology has attracted increasing attention from both academia and industry.With the growing complexity of tasks, researchers have found that reliance on smartphones' camera feed can be a limiting factor and reported several challenges for the agents, e.g., unfamiliarity with PVI's physical surroundings, difficulty in tracking moving objects, and interpreting various information in real-time. Since these challenges are related to understanding 3D environments and semantics from the live video, they can be potentially addressed by computer vision (CV) technologies. This dissertation aims to develop CV methods to enhance RSA services. The study has been carried out in three phases.In the first phase, we conducted a literature review and interviewed twelve RSA users to thoroughly understand the challenges in RSA. We summarized a comprehensive list of challenges, analyzed the causes and the technological needs, and thus identified some major CV problems in RSA. The second phase of the study is developing new methods to address key issues in the identified CV problems. In particular, we focus on addressing four representative CV problems: predicting the trajectories of pedestrians, recognizing persons or objects from camera feed, expanding the field of view of camera feed, and estimating the scale and depth in dual camera feed. The third phase is developing user-friendly RSA prototypes and verifying the CV methods through user studies. In our study, we focus on the exemplar problem of 3D indoor mapping and real-time localization. Specifically, we created an RSA prototype with ARKit-based iOS app for the third-phase study. Localizing RSA users on 3D maps in real-time has been proved effective for enhancing RSA services in our study. Through the three-phase study, we identified the challenges and CV problems in RSA, developed new methods for representative CV problems, and verified the effectiveness of exemplar CV techniques for supporting RSA.
533 $a Electronic reproduction. $b Ann Arbor, Mich. : $c ProQuest, $d 2024
538 $a Mode of access: World Wide Web
650 4 $a Information science. $3 561178
650 4 $a Electrical engineering. $3 596380
650 4 $a Computer engineering. $3 569006
650 4 $a Visualization. $3 574210
653 $a Remote sighted assistance
653 $a Computer vision technologies
653 $a Indoor mapping
655 7 $a Electronic books. $2 local $3 554714
690 $a 0544
690 $a 0464
690 $a 0723
710 2 $a The Pennsylvania State University. $3 845556
710 2 $a ProQuest Information and Learning Co. $3 1178819
773 0 $t Dissertations Abstracts International $g 85-05B.
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=30720551 $z click for full text (PQDT)