http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
- 中文 을 입력하시려면 zhongwen을 입력하시고 space를누르시면됩니다.
- 北京 을 입력하시려면 beijing을 입력하시고 space를 누르시면 됩니다.
RISS 인기검색어
- 검색키워드
- 저자 : ( Ming Yao Lim ) (검색결과 2 건)
- 무료
- 기관 내 무료
- 유료
-
Hedging Deep Features for Visual Tracking
Qi, Yuankai,Zhang, Shengping,Qin, Lei,Huang, Qingming,Yao, Hongxun,Lim, Jongwoo,Yang, Ming-Hsuan IEEE 2019 IEEE transactions on pattern analysis and machine Vol.41 No.5
<P>Convolutional Neural Networks (CNNs) have been applied to visual tracking with demonstrated success in recent years. Most CNN-based trackers utilize hierarchical features extracted from a certain layer to represent the target. However, features from a certain layer are not always effective for distinguishing the target object from the backgrounds especially in the presence of complicated interfering factors (e.g., heavy occlusion, background clutter, illumination variation, and shape deformation). In this work, we propose a CNN-based tracking algorithm which hedges deep features from different CNN layers to better distinguish target objects and background clutters. Correlation filters are applied to feature maps of each CNN layer to construct a weak tracker, and all weak trackers are hedged into a strong one. For robust visual tracking, we propose a hedge method to adaptively determine weights of weak classifiers by considering both the difference between the historical as well as instantaneous performance, and the difference among all weak trackers over time. In addition, we design a Siamese network to define the loss of each weak tracker for the proposed hedge method. Extensive experiments on large benchmark datasets demonstrate the effectiveness of the proposed algorithm against the state-of-the-art tracking methods.</P>
-
( Nai-chien Huan ),( Inn Shih Khor ),( Hema Yamini Ramarmuty ),( Ming Yao Lim ),( Kai Choon Ng ),( Alfieyanto Syaripuddin ),( Qin Zhi Lee ),( Wee Jing Teo ),( Kunji Kannan Sivaraman Kannan ) 대한결핵 및 호흡기학회 2020 대한결핵 및 호흡기학회 추계학술대회 초록집 Vol.128 No.-
Introduction Pleural fluid adenosine deaminase (pfADA) is a simple, rapid and inexpensive surrogate marker for tuberculous pleural effusion (TPE). A nationwide cut-off of 40U/L is currently used based on overseas data. There is a need to optimise the diagnostic utility of pfADA by establishing a local cut-off value. In this study, we aimed to: describe the demographics and clinical characteristics of patients with TPE and non-TPE; to determine the sensitivity and specificity of current pfADA of 40U/L; and to establish a new local pfADA cut-off for TPE. Methods We conducted a single centre, observational, prospective study of patients with exudative pleural effusion and pfADA measured from 1st October 2019 to 30th April 2020 at Queen Elizabeth Hospital, Malaysia. Results The diagnosis of analysed patients (n=93) included TPE (n=41), malignancy (n=28), parapneumonic effusion (n=12) and other causes (n=12) (figurer 1). The mean pfADA was 51.15 (SD=13.77)U/L among TPE group and 18.86 (SD=12.33)U/L among non- TPE. When analysis was restricted to TPE patients, the local pfADA cut-off is 29.6U/L, with sensitivity of 97.6% and specificity of 90.4% (figure 2). The current pfADA of 40U/L has a sensitivity of 87.8% and specificity of 92.3%. Conclusion We established a local pfADA cut-off of 29.6U/L for TPE. Optimising the utility of pfADA helps to enhance clinicians’ treatment confidence of TPE when initial work-up were inconclusive.
연관 검색어 추천
이 검색어로 많이 본 자료
활용도 높은 자료
