메뉴 건너뛰기
소속 기관 / 학교 인증
인증하면 논문, 학술자료 등을  무료로 열람할 수 있어요.
한국대학교, 누리자동차, 시립도서관 등 나의 기관을 확인해보세요
(국내 대학 90% 이상 구독 중)
고객센터 ENG
주제분류

논문 기본 정보

저자정보
(Hanyang University) (Hanyang University) (California State University) (Hanyang University) (Hanyang University) (Hanyang University)
저널정보
한국HCI학회 한국HCI학회 학술대회 PROCEEDINGS OF HCI KOREA 2022 학술대회 발표 논문집
오류 신고하기

피인용 0

검색

    초록·키워드

    Transportation mode detection (TMD) can help improving our daily life by understanding the mobility patterns of people. Such enhanced understanding of mobility can also be beneficial to people with mobility disabilities including wheelchair users. Moreover, as it is hard to collect large dataset from wheelchair users, it is important for TMD for wheelchair users (wTMD) to maintain similar performance on data from different period which might have dissimilar environmental or behavioral characteristics. However, we could not find studies evaluating such period independency of the wTMD model. Thus, we investigated wTMD performance on data from different period, and improved such performance by suggesting a new wTMD model. Our results showed that both our proposed model (DenseNet-based-model) and the baseline model (CNN-based-model) had period dependency, but our proposed model outperformed the baseline model when evaluated with data from different period. Our findings indicate the importance of the evaluation method and show that deep convolutional network with high information interchange can help improving wTMD performance on data from different period.

    최근 본 자료 전체보기