Browsing by Author "Chen, Yi-Wen"

Now showing 1 - 16 of 16

A Design for Integrated Face and Facial Expression Recognition
(2011)
Song, Kai-Tai
;
Chen, Yi-Wen
;
電機工程學系
;
Department of Electrical and Computer Engineering
An integrated face and facial expression recognition system has been designed and tested for robotic applications. Facial images from a web camera are first acquired for facial shape and texture model generation using active appearance model (AAM). Modified Lucas-Kanade image alignment algorithm was adopted to find facial features as well as the texture model of AAM to construct facial texture parameters. These parameters are used to train back propagation neural networks(BPNN) for face and facial expression recognition. A novel design is proposed for an integrated facial expression recognition system. In the first stage, face recognition is performed to find user's identity; then the facial-expression database of the recognized user is employed to recognize his/her facial expressions. Experimental result based on BU-3DFE database show that a face recognition rate of 98.3% is achieved. The facial expression recognition rate of the proposed integrated method (using personal facial expression classifiers) is 83.8%, an improvement compared with 69.6% of using conventional classifiers.
A Parametric Window Design for OBMC with Variable Block Size Motion Estimates
(2009)
Chen, Yi-Wen
;
Wang, Tse-Wei
;
Tseng, Yu-Cheng
;
Peng, Wen-Hsiao
;
Lee, Suh-Yin
This paper addresses the problem of adapting overlapped block motion compensation (OBMC) windows for use with variable blocksize motion estimates. We tackle the problem by using a parametric window design, which expresses, based on a statistical motion model, the optimal weights as a function of the distances between the predicted pixel and its nearby block centers. The formula enables both prediction weights and prediction order to be adapted on a pixel-by-pixel basis. Extensive experiments have been conducted using JM 12.4. Compared with conventional block motion compensation, our scheme shows a bitrate saving of 18%, (5% on average) while maintaining the same or even higher PSNR (0.1 dB). It also provides a competitive advantage to variable block size motion compensation. Additionally, a hybrid of the two techniques achieves a further bitrate reduction of 13%. The result, nevertheless, provides only a lower bound on what is achievable since both motion estimation and mode decision were accomplished without considering OBMC. Further improvement is expected by incorporating iterative methods.
A Synthesis-Quality-Oriented Depth Refinement Scheme for MPEG Free Viewpoint Television (FTV)
(2009)
Chen, Chun-Chi
;
Chen, Yi-Wen
;
Yang, Fu-Yao
;
Peng, Wen-Hsiao
;
資訊工程學系
This paper addresses the problem of refining depth information front the received reference and depth images within the MPEG FT V framework. An analytical model is first developed to approximate the per-pixel synthesis distortion (caused by depth-image compression) as a function of depth-error variances, intensity variations, ground-truth depth and virtual camera locations. We then follow the model to detect unreliable depth pixels by inspecting intensity gradients and to refine their values with a candidate-based block disparity search. Additional side information is transmitted to make both operations robust against compression effects. Experimental results show that our scheme offers an average PSNR improvement of 1.2 dB over MPEG FTV and consistently outperforms the state-of-the-art methods. Moreover, it can remove synthesis artifacts to a great extent, producing a result that is very close in appearance to the ground-truth view image.
Ambient-light-adaptive image quality enhancement for full-color e-paper displays using a saturation-based tone-mapping method
(2018-03-01)
Qin, Zong
;
Chen, Yi-Wen
;
Lin, Fang-Cheng
;
Hung, Chi-Mao
;
Shieh, Han-Ping David
Full-color e-paper displays that have a small color gamut encounter image tone distortions and contour artifacts in high-saturation regions. To solve these issues, in this paper, we comprehensively measure the colorimetric responses of input image signals and demonstrate that the perceptual colorfulness in terms of chroma abnormally declines when the input saturation is beyond a certain value. Accordingly, saturation-based tone-mapping curves are developed to map the abnormal high saturation to the range that the display is able to normally render. By testing several test charts and natural images, the recoveries of the original image tones and the removals of the contour artifacts are experimentally verified. More important, by discussing how to conduct a database of tone-mapping curves for different ambient lights and utilize the database for an unknown ambient light, the proposed method is proven to be completely ambient light adaptive.
BI-PREDICTION COMBINING TEMPLATE AND BLOCK MOTION COMPENSATIONS
(2011)
Lee, Chung-Lin
;
Chen, Chun-Chi
;
Chen, Yi-Wen
;
Wu, Mu-Hsuan
;
Wu, Chung-Hao
This paper introduces a bi-prediction scheme with only a motion overhead as for unidirectional prediction. It combines motion vectors found by template and block matchings with the overlapped block motion compensation (OBMC). An optimal window function is designed based on a model-based framework. Additionally, the concept of adaptive motion merging is incorporated to enable a template-matching-free implementation. Three algorithms featuring different performance and complexity trade-offs are implemented using the TMuC-0.9_HM software and tested with the common test conditions. Relative to the anchor, the best of them achieves an average BD-rate saving of 2.2%, with a minimum of 0.2% and a maximum of 4.7%.
Content-aware fast motion estimation algorithm
(2008-05-01)
Chen, Yi-Wen
;
Hsiao, Ming-Ho
;
Chen, Hua-Tsung
;
Liu, Chi-Yu
;
Lee, Suh-Yin
In this paper, we propose the Content-Aware Fast Motion Estimation Algorithm (CAFME) that can reduce computation complexity of motion estimation (ME) in H.264/AVC while maintaining almost the same coding efficiency. Motion estimation can be divided into two phases: searching phase and matching phase. In searching phase.. we propose the Simple Dynamic Search Range Algorithm (SDSR) based on video characteristics to reduce the number of search points (SP). In matching phase, we integrate the Successive Elimination Algorithm (SEA) and the integral frame to develop a new SEA for H.264/AVC video compression standard, called Successive Elimination Algorithm with Integral Frame (SEAIF). Besides, we also propose the Early Termination Algorithm (ETA) to early terminate the motion estimation of current block. We implement the proposed algorithm in the reference software JM9.4 of H.264/AVC and the experimental results show that our proposed algorithm can reduce the number of search points about 93.1%, encoding time about 42%, while maintaining almost the same bitrate and PSNR. (c) 2008 Elsevier Inc. All rights reserved.
Content-aware video adaptation under low-bitrate constraint
(2007)
Hsiao, Ming-Ho
;
Chen, Yi-Wen
;
Chen, Hua-Tsung
;
Chou, Kuan-Hung
;
Lee, Suh-Yin
With the development of wireless network and the improvement of mobile device capability, video streaming is more and more widespread in such an environment. Under the condition of limited resource and inherent constraints, appropriate video adaptations have become one of the most important and challenging issues in wireless multimedia applications. In this paper, we propose a novel content- aware video adaptation in order to e. effectively utilize resource and improve visual perceptual quality. First, the attention model is derived from analyzing the characteristics of brightness, location, motion vector, and energy features in compressed domain to reduce computation complexity. Then, through the integration of attention model, capability of client device and correlational statistic model, attractive regions of video scenes are derived. The information object- ( IOB-) weighted rate distortion model is used for adjusting the bit allocation. Finally, the video adaptation scheme dynamically adjusts video bitstream in frame level and object level. Experimental results validate that the proposed scheme achieves better visual quality effectively and effciently.
Effect of NH(3) Plasma Nitridation on Hot-Carrier Instability and Low-Frequency Noise in Gd-Doped High-kappa Dielectric nMOSFETs
(2011-03-01)
Chen, Yu-Ting
;
Chen, Kun-Ming
;
Lin, Cheng-Li
;
Yeh, Wen-Kuan
;
Huang, Guo-Wei
The effects of post-NH(3) plasma nitridation on device's hot-carrier instability and low-frequency noise in the Hf-based high-kappa/metal-gate n-channel metal-oxide-semiconductor field-effect transistors (nMOSFETs) with gadolinium (Gd) cap layers are investigated. With postnitridation, the direct-current and 1/f noise characteristics can be improved apparently. Moreover, a hot-carrier stressing-induced threshold voltage shift can be also suppressed despite of a similar transconductance degradation when comparing with that in the device without nitridation. With the charge-pumping and low-frequency noise measurements, we find that the bulk-and interfacial-trap densities can be reduced with nitrogen incorporation. The reduction of bulk and interfacial traps can be contributed to the suppression of Gd diffusion into a high-kappa layer. In this paper, appropriate post-NH(3) plasma nitridation can improve the device performance and reliability and low-frequency noise for a gate-first high-kappa/metal-gate nMOSFET with a Gd cap layer.
Effect of NH3 Plasma Nitridation on Hot-Carrier Instability and Low-Frequency Noise in Gd-Doped High-kappa Dielectric nMOSFETs
(2011-03-01)
Chen, Yu-Ting
;
Chen, Kun-Ming
;
Lin, Cheng-Li
;
Yeh, Wen-Kuan
;
Huang, Guo-Wei
The effects of post-NH3 plasma nitridation on device's hot-carrier instability and low-frequency noise in the Hf-based high-kappa/metal-gate n-channel metal-oxide-semiconductor field-effect transistors (nMOSFETs) with gadolinium (Gd) cap layers are investigated. With postnitridation, the direct-current and 1/f noise characteristics can be improved apparently. Moreover, a hot-carrier stressing-induced threshold voltage shift can be also suppressed despite of a similar transconductance degradation when comparing with that in the device without nitridation. With the charge-pumping and low-frequency noise measurements, we find that the bulk-and interfacial-trap densities can be reduced with nitrogen incorporation. The reduction of bulk and interfacial traps can be contributed to the suppression of Gd diffusion into a high-kappa layer. In this paper, appropriate post-NH3 plasma nitridation can improve the device performance and reliability and low-frequency noise for a gate-first high-kappa/metal-gate nMOSFET with a Gd cap layer.
ON THE ANALYSIS AND DESIGN OF MOTION SAMPLING STRUCTURE FOR ADVANCED MOTION-COMPENSATED PREDICTION
(2010)
Tseng, Yu-Chen
;
Wu, Chung-Hao
;
Chen, Yi-Wen
;
Wang, Tse-Wei
;
Peng, Wen-Hsiao
This paper addresses the problem of improving motion sampling efficiency for motion-compensated prediction (MCP). We provide a theoretical framework for analyzing the effect of motion sampling structure on MCP efficiency. It is shown that the sampling grid induced by the quadtree partition in H.264/AVC is suboptimal. To improve sampling efficiency, we propose a new pattern, which provides sampling points at both the center and the top-left corner of a macroblcok. When contrasted with conventional NxN/2 block partition, the proposed scheme performs consistently and significantly better in subjective and objective quality.
Parametric OBMC for Pixel-Adaptive Temporal Prediction on Irregular Motion Sampling Grids
(2012-01-01)
Chen, Yi-Wen
;
Peng, Wen-Hsiao
;
資訊工程學系
;
Department of Computer Science
This paper adapts overlapped block motion compensation (OBMC) to suit variable block-size motion partitioning. The motion vectors (MVs) for various partitions are formalized as motion samples taken on an irregular grid. From this viewpoint, determining OBMC weights to associate with these samples becomes an under-determined problem since a distinct solution has to be sought for each prediction pixel. In this paper, we tackle this problem by expressing the optimal weights in closed form based on parametric signal assumptions. In particular, the computation of this solution requires only the geometric relations between the prediction pixel and its nearby block centers, leading to a generic framework capable of reconstructing temporal predictors from any irregularly sampled MVs. A modified implementation is also proposed to address the MV location uncertainty and to reduce computational complexity. Experimental results demonstrate that our scheme performs better than similar previous works, and when compared to the recently proposed Quadtree-based adaptive loop filter and enhanced adaptive interpolation filter, show a comparable gain. Furthermore, the combination of it with either of them gives a combined effect that is almost the sum of their separate improvements.
Physics-based ball tracking and 3D trajectory reconstruction with applications to shooting location estimation in basketball video
(2009-04-01)
Chen, Hua-Tsung
;
Tien, Ming-Chun
;
Chen, Yi-Wen
;
Tsai, Wen-Jiin
;
Lee, Suh-Yin
The demand for computer-assisted game study in sports is growing dramatically. This paper presents a practical video analysis system to facilitate semantic content understanding. A physics-based algorithm is designed for ball tracking and 3D trajectory reconstruction in basketball videos and shooting location statistics can be obtained. The 2D-to-3D inference is intrinsically a challenging problem due to the loss of 3D information in projection to 2D frames. One significant contribution of the proposed system lies in the integrated scheme incorporating domain knowledge and physical characteristics of ball motion into object tracking to overcome the problem of 2D-to-3D inference. With the 2D trajectory extracted and the camera parameters calibrated, physical characteristics of ball motion are involved to reconstruct the 3D trajectories and estimate the shooting locations. Our experiments on broadcast basketball videos show promising results. We believe the proposed system will greatly assist intelligence collection and statistics analysis in basketball games. (C) 2008 Elsevier Inc. All rights reserved.
Preliminary clinical and radiographic outcomes of proximal humeral fractures: comparison of ALPS and PHILOS plating in Asian patients in Taiwan
(2020-08-28)
Chen, Chun-Yen
;
Chang, Hao-Wei
;
Hsieh, Shang-Lin
;
Chang, Chien-Chun
;
Tsai, Chun-Hao
Background Up to 20% of proximal humeral fractures need to be treated operatively. However, numerus complications were reported by using fixed angled locking plates. The ALPS Proximal Humerus Plating System is a new design implant with novel design features. The aim of this study was to compare the preliminary clinical outcomes and complications of proximal humeral fractures treated with either ALPS or the proximal humeral internal locking system (PHILOS) in Asian patients in Taiwan. Methods Between January 2016 and December 2018, 66 patients with displaced proximal humeral fractures were analyzed retrospectively, of whom 31 underwent ALPS implant treatment and 35 underwent PHILOS implant treatment. Intraoperative blood loss and operation time, postoperative Constant-Murley Shoulder Outcome (Constant-Murley) score, and complications variables were recorded for the comparison. All cases were regularly followed up for at least 1 year. Results The mean follow-up period was 400.8 days (range, 367-446 days). Union was achieved in 98.5% of patients (65/66). The ALPS group yielded similar radiologic and clinical outcomes to the PHILOS plating group for treating displaced proximal humeral fractures, including operation time, intraoperative blood loss, the Constant-Murley score, and varus malunion (P> 0.05, respectively). However, the incidence of total postoperative complications in the ALPS group was significantly lower than in the PHILOS group (P< 0.05). There was a trend of a lower complication rate of screws/pegs protrusion, avascular necrosis, subacromial impingement, postoperative infection, and reoperation in the ALPS group, although it was not statistically significant (P> 0.05, respectively). Conclusion The ALPS group yielded similar radiologic and clinical outcomes to the PHILOS plating group for displaced proximal humeral fractures, but the ALPS group had a significantly lower total rate of complications. Therefore, ALPS may be a better option for treating proximal humeral fractures. Further larger clinical studies are needed to confirm the findings presented here.
Shot classification of basketball videos and its application in shooting position extraction
(2007)
Tien, Min-Chun
;
Chen, Hua-Tsung
;
Chen, Yi-Wen
;
Hsiao, Ming-Ho
;
Lee, Suh-Yin
In this paper, we propose a system that can automatically segment a basketball video into several clips on the basis of a GOP-based scene change detection method. The length of each clip and the number of dominant color pixels of each frame are used to classify shots into close-up view, medium view, and full court view. Full court view shots are chosen to do advanced analyses such as ball tracking and parameter extracting for the transformation from a 3D real-world court to a 2D image. After that, we map points in the 2D image to the corresponding coordinates in a real-world court by some physical properties of the 3D shooting trajectory, and compute the statistics of all shooting positions. Eventually we can obtain the information about the most possible shooting positions of a professional basketball team, which is useful for opponents to adopt appropriate defense tactics.
User Identification Design by Fusion of Face Recognition and Speaker Recognition
(2012-01-01)
Lin, Chao-Yu
;
Song, Kai-Tai
;
Chen, Yi-Wen
;
Chien, Shuo-Cheng
;
Chen, Sin-Horng
Intelligent human-robot interface helps a mobile robot to extract external information and interact with a user. User identification information allows a robot to generate appropriate behaviors and make personalized human-robot interaction (PHRI) more natural and safe. Most of service robots move around in various application settings, and biometric features always change during relative movements. In this work, a fusion recognition system based on face and speaker classifiers is designed to achieve the user recognition in various environments. We propose a method to fuse face and speech recognition results based on the concept of confidence indices. Through the derived confidence indices, the system combines the two classifiers and gives improved recognition results. Experiments on a laboratory home service robot show that person recognition rates in 15 environmental situations all outperform those of using single modality.
VOSTR: Video Object Segmentation via Transferable Representations
(1970-01-01)
Chen, Yi-Wen
;
Tsai, Yi-Hsuan
;
Lin, Yen-Yu
;
Yang, Ming-Hsuan
;
交大名義發表
In order to learn video object segmentation models, conventional methods require a large amount of pixel-wise ground truth annotations. However, collecting such supervised data is time-consuming and labor-intensive. In this paper, we exploit existing annotations in source images and transfer such visual information to segment videos with unseen object categories. Without using any annotations in the target video, we propose a method to jointly mine useful segments and learn feature representations that better adapt to the target frames. The entire process is decomposed into three tasks: (1) refining the responses with fully-connected CRFs, (2) solving a submodular function for selecting object-like segments, and (3) learning a CNN model with a transferable module for adapting seen categories in the source domain to the unseen target video. We present an iterative update scheme between three tasks to self-learn the final solution for object segmentation. Experimental results on numerous benchmark datasets demonstrate that the proposed method performs favorably against the state-of-the-art algorithms.