東華大學圖書館 |

Language: English

Help

回圖書館首頁

手機版館藏查詢

Back

Switch To: Labeled | MARC Mode | ISBD

Computer Vision Methods for Enhancin...

Yu, Rui.

Linked to FindBook

Google Book

Amazon

博客來

Computer Vision Methods for Enhancing Remote Sighted Assistance.

Record Type:	Electronic resources : Monograph/item
Title/Author:	Computer Vision Methods for Enhancing Remote Sighted Assistance./
Author:	Yu, Rui.
Published:	Ann Arbor : ProQuest Dissertations & Theses, : 2023,
Description:	177 p.
Notes:	Source: Dissertations Abstracts International, Volume: 85-05, Section: B.
Contained By:	Dissertations Abstracts International85-05B.
Subject:	Visualization. -
Online resource:	https://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=30720551
ISBN:	9798380730495

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

Computer Vision Methods for Enhancing Remote Sighted Assistance. - Ann Arbor : ProQuest Dissertations & Theses, 2023 - 177 p.

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

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

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.

ISBN: 9798380730495Subjects--Topical Terms:

586179
Visualization.
Subjects--Index Terms:

Remote sighted assistance

Computer Vision Methods for Enhancing Remote Sighted Assistance.
LDR:03473nmm a2200373 4500 001 2400080
005 20240924101856.5
006 m o d
007 cr#unu||||||||
008 251215s2023 ||||||||||||||||| ||eng d
020 $a 9798380730495
035 $a (MiAaPQ)AAI30720551
035 $a (MiAaPQ)PennState_23128rzy54
035 $a AAI30720551
040 $a MiAaPQ $c MiAaPQ
100 1 $a Yu, Rui. $3 3184433
245 1 0 $a Computer Vision Methods for Enhancing Remote Sighted Assistance.
260 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2023
300 $a 177 p.
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.
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.
590 $a School code: 0176.
650 4 $a Visualization. $3 586179
650 4 $a Computer engineering. $3 621879
650 4 $a Electrical engineering. $3 649834
650 4 $a Information science. $3 554358
653 $a Remote sighted assistance
653 $a Computer vision technologies
653 $a Indoor mapping
690 $a 0544
690 $a 0464
690 $a 0723
710 2 $a The Pennsylvania State University. $3 699896
773 0 $t Dissertations Abstracts International $g 85-05B.
790 $a 0176
791 $a Ph.D.
792 $a 2023
793 $a English
856 4 0 $u https://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=30720551