Browsing by Author "Chen, Chun-Min"

Now showing 1 - 7 of 7

A novel approach for semantic event extraction from sports webcast text
(2014-08-01)
Chen, Chun-Min
;
Chen, Ling-Hwei
;
資訊工程學系
;
Department of Computer Science
Semantic event extraction is helpful for video annotation and retrieval. For sports video, most previous works detect events by video content itself. Some useful external knowledge has been researched recently. In this paper, we proposed an unsupervised approach to extract semantic events from sports webcast text. First, unrelated words in the descriptions of webcast text are filtered out, and then the filtered descriptions are clustered into significant event categories. Finally, the keywords for each event category are extracted. According to our experimental results, the proposed approach actually extracts significant text events, which can be used for further video indexing and summarization. Furthermore, we also provide a hierarchical searching scheme for text event retrieval.
A novel approach for semantic event extraction from sports webcast text
(2014-08-01)
Chen, Chun-Min
;
Chen, Ling-Hwei
;
資訊工程學系
;
Department of Computer Science
Semantic event extraction is helpful for video annotation and retrieval. For sports video, most previous works detect events by video content itself. Some useful external knowledge has been researched recently. In this paper, we proposed an unsupervised approach to extract semantic events from sports webcast text. First, unrelated words in the descriptions of webcast text are filtered out, and then the filtered descriptions are clustered into significant event categories. Finally, the keywords for each event category are extracted. According to our experimental results, the proposed approach actually extracts significant text events, which can be used for further video indexing and summarization. Furthermore, we also provide a hierarchical searching scheme for text event retrieval.
A novel method for slow motion replay detection in broadcast basketball video
(2015-11-01)
Chen, Chun-Min
;
Chen, Ling-Hwei
;
資訊工程學系
;
Department of Computer Science
Slow motion replays are valuable for sports video analysis. Many methods for slow motion replay detection have been proposed, and they are classified into two categories. One assumes that a replay is sandwiched by a pair of visually similar special digital video effects, but the assumption is not always true in basketball videos. The other analyzes replay features to distinguish replay segments from non-replay segments. The results are not satisfactory since some features (e.g. dominant color of sports field) are not applicable for basketball. Most replay detectors focus on soccer videos. In this paper, we propose a novel idea to detect slow motion replays in basketball videos. The existence of scoreboard is referred to filter large amount of non-replay frames, this improves detection accuracy. After scoreboard frame filtering, every consecutive non-scoreboard frame sequence bounded by scoreboard frames are considered as a non-scoreboard segment. Characteristics of replays and non-replays are observed to create features, which can be used to detect replays and prune non-replays from non-scoreboard segments. Experimental results show that the proposed replay detection method is applicable for both kinds of basketball videos with/without TV commercials. As compared with previous researches for basketball videos, our method presents the superior performance.
Anatomic Differences between Parkinson\'s Disease and Essential Tremor Using ICA-based Brain Morphometry
(2013)
Duann, Jeng-Ren
;
Lin, Ching-Hung
;
Chen, Chun-Min
;
Lu, Ming-Kuei
;
Tsai, Chon-Haw
This study used independent component analysis (ICA) to decompose the surrogate image data by concatenating in space multiple subjects\' anatomical images after spatial normalization and smoothing to make all images in the same space. The multiple-subject anatomical image data included three different subject/patient populations, namely the Parkinson\'s disease (PD) and essential tremor (ET) as well as age-matched normal control (NC) subjects. In so doing, we were to extract the independent components showing significant group differences caused by the morphometric changes due to the neurological diseases. Such an independent component-based morphometry (ICBM) applied to both the grouped gray-matter and white-matter image data could be used to successfully reveal the brain areas within the gray and white matters showing significant group differences. In gray matter, we found a majority of basal ganglia brain areas associated with NC > PD, which well represented the causes of PD reported in the literature. In addition, the ICBM also found the brain areas of right superior temporal gyrus, right angular gyrus (BA 39), right inferior temporal gyrus, and left middle temporal gyrus (BA 37) associated with NC > ET. On the other hand, a more complicated patterns were found in the white matter associated with PD > ET and ET > PD conditions.
NOVEL FRAMEWORK FOR SPORTS VIDEO ANALYSIS: A BASKETBALL CASE STUDY
(2014)
Chen, Chun-Min
;
Chen, Ling-Hwei
;
資訊工程學系
;
Department of Computer Science
Semantic event and slow motion replay extraction for sports videos have become hot research topics. Most researches analyze every video frame; however, semantic events only appear in frames with scoreboard, whereas replays only appear in frames without scoreboard. Extracting events and replays from unrelated frames causes defects and leads to degradation of performance. In this paper, a novel framework is proposed to tackle challenges of basketball video analysis. In the framework, a scoreboard detector is first provided to divide video frames to two classes, with/without scoreboard. Then, a semantic event extractor is presented to extract semantic events from frames with scoreboard and a slow motion replay extractor is proposed to extract replays from frames without scoreboard. Experimental results show that the proposed framework is practicable for basketball videos. It is expected that the proposed framework can be extended to other sports.
Older patients' depressive symptoms 6 months after prolonged hospitalization: Course and interrelationships with major associated factors
(2014-05-01)
Chen, Chun-Min
;
Huang, Guan-Hua
;
Chen, Cheryl Chia-Hui
;
統計學研究所
;
Institute of Statistics
The aim of this study was to examine the course of depressive symptoms in older patients 6 months following a prolonged, acute hospitalization, especially the interrelationships among depressive symptoms and its major associated factors. For this study, we conducted a secondary analysis of data from a prospective cohort study of 351 patients aged 65 years and older. Participants were recruited from five surgical and medical wards at a tertiary medical center in northern Taiwan and assessed at three time points: within 48 h of admission, before discharge, and 6 months post-discharge. The course of depressive symptoms was dynamic with symptoms increased spontaneously and substantially during hospitalization and subsided at 6 months after discharge, but still remained higher than at admission. Overall, 26.7% of older patients at hospital discharge met established criteria for minor depression (15-item Geriatric Depressive Scale (GDS-15) scores 5-9) and 21.2% for major depression (GDS-15 scores >10). As the strongest associated factors, functional dependence and nutritional status influenced depressive symptoms following hospitalization. Depressive symptoms at discharge showed significant cross-lagged effects on functional dependence and nutritional status at 6 months after discharge, suggesting a reciprocal, triadic relationship. Thus, treating one condition might improve the other. Targeting the triad of depressive symptoms, functional dependence, and nutritional status, therefore, is essential for treating depressive symptoms and improving the overall health of older adults hospitalized for acute illness. (C) 2014 Elsevier Ireland Ltd. All rights reserved.
Pre-surgical Geriatric Syndromes, Frailty, and Risks for Postoperative Delirium in Older Patients Undergoing Gastrointestinal Surgery: Prevalence and Red Flags
(2015-05-01)
Chen, Cheryl Chia-Hui
;
Lin, Ming-Tsan
;
Liang, Jin-Tung
;
Chen, Chun-Min
;
Yen, Chung-Jen
We sought to identify and evaluate red flags for pre-surgical geriatric conditions (geriatric syndromes, frailty, and risks for postoperative delirium) in older patients undergoing gastrointestinal surgery. Older individuals (a parts per thousand yen65 years) undergoing major elective gastrointestinal surgery from 2009 to 2012 were enrolled and assessed preoperatively. Participants (N = 379; mean age = 74.5 +/- 5.9 years) primarily underwent colorectal (54.3 %), gastric (21.9 %), and pancreatobiliary (12.6 %) surgery. Overall, 30.9 % had existing geriatric syndromes, 26.7 % were frail, and 22.8 % had > 3 risk factors for postoperative delirium. The largest proportion (45.7 %) presented with at least one geriatric condition. Patients with or without geriatric conditions were discriminated with adequate sensitivity (67 %), specificity (84 %), and positive predictive value (77 %) by eight red flags: age a parts per thousand yen75 years (OR, 2.86; P < 0.001), eating soft food (OR, 3.63; P = 0.001), reported hypertension (OR, 2.8; P = 0.001), weight loss > 3 kg (OR, 4.79; P < 0.001), fair-to-weak grip strength (OR, 2.53; P = 0.001), sleeplessness (OR, 2.57; P = 0.001), no-better-than-peer perceived health (OR, 1.88; P = 0.022), and short-term inability to recall two of three common words (OR, 1.81; P = 0.025). Eight red flags covered as part of history and physical examination are well suited to screen patients for geriatric conditions indicating the need for preoperative geriatric assessments and optimization.