Browsing by Author "Wu, Tung-Yu"

Now showing 1 - 11 of 11

A 1.86mJ/Gb/Query Bit-Plane Payload Machine Learning Processor in 90nm CMOS
(2018-01-01)
Ku, Fang-Ju
;
Wu, Tung-Yu
;
Liao, Yen-Chin
;
Chang, Hsie-Chia
;
Wong, Wing Hung
This paper presents an implementation of an energy efficient hit-plane payload design for machine learning processor. The proposed architecture facilitates high parallelism and high data bandwidth and thus improves the model learning/training time of machine learning algorithms. By assembling multiple bits as a bit-plane and enlarging query parallelism with a central compare-Hag updater, data processing parallelism can be increased. Binary sequential partition (BSP), a fast density estimation algorithm capable of dealing with high dimensional data sets, is realized. Fabricated in 90nm IP9M CMOS process, the processing rate can achieve 16.9 Gb/sec with 8 queries for data dimension D=210. The test chip integrates 64 counting cells and provides 5 modes with power consumptions of 1.86mJ/Gb per Query.
CONFNET: PREDICT WITH CONFIDENCE
(2018-01-01)
Wan, Sheng
;
Wu, Tung-Yu
;
Wong, Wing H.
;
Lee, Chen-Yi
;
電子工程學系及電子研究所
In this paper, we propose Confidence Network (ConfNet) which not only makes predictions on input images but also generates a confidence score that estimates the probability of correctness of each prediction. Furthermore, Confidence Loss is proposed to make ConfNet automatically learn confidence scores in the training phase. The experiments on two public datasets show that the confidence scores generated by ConfNet are highly correlated with the model accuracy and outperforms two related methods. When stacking two ConfNets in a cascade structure, 3.8x computational cost can be saved compared to the single state-of-the-art model with only 0.1% increase of error rate.
CORRELATION-BASED FACE DETECTION FOR RECOGNIZING FACES IN VIDEOS
(2018-01-01)
Hsu, Heng-Wei
;
Wu, Tung-Yu
;
Wong, Wing Hung
;
Lee, Chen-Yi
;
電子工程學系及電子研究所
Finding the locations and identities of faces in videos is a very important task in numerous applications. In this paper, we propose a correlation-based face detection approach to improve the performance of face recognition tasks for videos. We apply correlation measures to pairs of response maps which are generated from automatically selected neurons in deep convolutional neural network (CNN) models to detect faces in each video frame. The embeddings extracted from faces cropped by our proposed approach are more consistent across each video sequence and more suitable for face recognition and clustering tasks. Experimental results from the YouTube Faces (YTF) dataset demonstrate that our proposed approach is more robust and achieves better recognition accuracy compared to state-of-the-art face detection approaches.
Cross Layer Adaptation with QoS Guarantees for Wireless Scalable Video Streaming
(2012-09-01)
Lin, Hsuan-Li
;
Wu, Tung-Yu
;
Huang, Ching-Yao
;
電機工程學系
;
Department of Electrical and Computer Engineering
In this paper, a cross-layer adaptation scheme is proposed for quality of service (QoS) provision in the scalable video streaming of high definition (HD) content. The cross-layer parameters, which contain the video rate, payload length of a packet, the mode of modulation and coding scheme (MCS), can be dynamically adapted to minimize distortion of a video streaming under the given delay bound. Based on the channel quality and rate-distortion parameters, the proposed scheme formulates the problem of parameter selection into an optimization problem. Simulation results show that our approach guarantees video quality under QoS constraints.
DIABETIC RETINOPATHY DETECTION BASED ON DEEP CONVOLUTIONAL NEURAL NETWORKS
(2018-01-01)
Chen, Yi-Wei
;
Wu, Tung-Yu
;
Wong, Wing-Hung
;
Lee, Chen-Yi
;
電子工程學系及電子研究所
Diabetic retinopathy is the primary cause of blindness in the working-age population of the developed world. Diagnosing the disease heavily relies on imaging studies, which is a time consuming and a manual process performed by trained clinicians. Enhancing the accuracy and speed of the detection process can potentially have a significant impact on population health via early diagnosis and intervention. Motivated by this, we propose a recognition pipeline based on deep convolutional neural networks. In our pipeline, we design lightweight networks called SI2DRNet-v1 along with six methods to further boost the detection performance. Without any fine-tuning, our recognition pipeline outperforms state of the art on the Messidor dataset along with 5.26x fewer in total parameters and 2.48x fewer in total floating operations.
IMPROVED SPECTRAL MATTING BY ITERATIVE K-MEANS CLUSTERING AND THE MODULARITY MEASURE
(2012)
Wu, Tung-Yu
;
Juan, Hung-Hui
;
Lu, Henry Horng-Shing
;
統計學研究所
;
Institute of Statistics
Spectral matting is a useful technique for image matting problem. A crucial issue of spectral matting is to determine the number of matting components which has large impacts on the matting performance. In this paper, we propose an improved framework based on spectral matting in order to solve this limitation. Iterative K-means clustering with the assistance of the modularity measure is adopted to obtain the hard segmentation that can be used as the initial guess of soft matting components. The number of matting components can be determined automatically because the improved framework will search possible image components by iteratively dividing image subgraphs.
Payload Length Adaptation for Wireless Video Transmission in Multicarrier Systems
(2011)
Lin, Hsuan-Li
;
Wu, Tung-Yu
;
Huang, Ching-Yao
;
電機工程學系
;
Department of Electrical and Computer Engineering
We present an optimization architecture for a cross-layer design in wireless video transmission and investigate the multicarrier effect on wireless communication. To transmit video frames with delay bound and target frame error rate (FER) constraint, we formulate an optimization problem to minimize the required resource units for a single user by adjusting payload length, M-ary modulation, block size, and code rate. Instead of exhaustive approach, we propose a minimal required resource units (MRRU) algorithm with feedback of mean mutual information per bit (MMIB) to resolve this problem by binomial distribution function approximation. Finally, the performance comparison between MRRU algorithm and maximum throughput approach is given, and the effect on multicarrier system in a correlated Rayleigh fading channel is shown.
QuatNet: Quaternion-Based Head Pose Estimation With Multiregression Loss
(2019-04-01)
Hsu, Heng-Wei
;
Wu, Tung-Yu
;
Wan, Sheng
;
Wong, Wing Hung
;
Lee, Chen-Yi
Head pose estimation has attracted immense research interest recently, as its inherent information significantly improves the performance of face-related applications such as face alignment and face recognition. In this paper, we conduct an in-depth study of head pose estimation and present a multiregression loss function, an L2 regression loss combined with an ordinal regression loss, to train a convolutional neural network (CNN) that is dedicated to estimating head poses from RGB images without depth information. The ordinal regression loss is utilized to address the nonstationary property observed as the facial features change with respect to different head pose angles and learn robust features. The L2 regression loss leverages these features to provide precise angle predictions for input images. To avoid the ambiguity problem in the commonly used Euler angle representation, we further formulate the head pose estimation problem in quaternions. Our quaternion-based multiregression loss method achieves state-of-the-art performance on the AFLW2000, AFLW test set, and AFW datasets and is closing the gap with methods that utilize depth information on the BIWI dataset.
Thiophene-Vinylene-Thiophene-Based Donor-Acceptor Copolymers with Acetylene-Inserted Branched Alkyl Side Chains To Achieve High Field-Effect Mobilities
(2018-11-13)
Chiou, De-Yang
;
Su, Yen-Chen
;
Hung, Kai-En
;
Hsu, Jhih-Yang
;
Hsu, Tze-Gang
2-Alkyl((1))alkyl((2))-type aliphatic side chains with a branching point position at the C-2-position (such as 2-ethylhexyl or 2-octayldodecyl) have been widely implanted into numerous donor-acceptor conjugated copolymers for solution processable transistors or organic solar cells. However, the tertiary branching site located at the second carbon inevitably imposes steric hindrance that twists the main-chain coplanarity and attenuates interchain interactions. In this research, we developed a new two-dimensonal thiophene-vinylene-thiophene (TVT) derivative where a carbon-carbon triple bond is inserted between the thiophene unit and the 2-octyldodecyl group. This acetylene-incorporated TVT (aTVT) was copolymerized with 5,10-di(thiophen-2-yl)naphtho[1,2-c:5,6-c']bis-([1,2,5]thiadiazole) (DTNT) and 5,6-difluoro-4,7-di(thiophen-2-yl[benzo][1,2,5]thiadiazole (DTFBT) to form the polymers PaTVT-NT and PaTVT-FBT, respectively. PTVT-FBT, without the triple bond, was also prepared for comparison. The insertion of a linear triple bond moves the tertiary carbon away from the main chain to reduce the steric hindrance, thereby improving the main-chain coplanarity and facilitating the interchain interactions. The acetylene-incorporated copolymers show better thermal stability, red-shifted absorption spectra, stronger intermolecular aggregation, lower-lying electron affinity, and much higher solid-state crystallinity. Due to the linear and coplanar polymeric backbone supported by theoretical calculation, PaTVT-NT exhibits high crystallinity and adopts strong stacking with an edge-on orientation in the thin film evidenced by 2D-GIXRD, leading to a high p-type OFET mobility up to 1.27 cm(2) V-1 s(-1) with an on-off ratio of 9.22 X 10(5). This value represents the highest value among the NT-based polymers. PaTVT-FBT also achieved a high mobility of 0.78 cm(2) which greatly outperforms the corresponding nonacetylene PTVT-FBT counterpart. Most importantly, the preparation of 2-alkyl((1))alkyl((2))-acetylenyl side chain is synthetically feasible, which can be easily applied to create new conjugated polymers for high-performance solution-processable optoelectronics.
Towards high performance data analytic on heterogeneous many-core systems: A study on Bayesian Sequential Partitioning
(2018-12-01)
Lai, Bo-Cheng
;
Wu, Tung-Yu
;
Chiu, Tsou-Han
;
Li, Kun-Chun
;
Lee, Chia-Ying
Bayesian Sequential Partitioning (BSP) is a statistically effective density estimation method to comprehend the characteristics of a high dimensional data space. The intensive computation of the statistical model and the counting of enormous data have caused serious design challenges for BSP to handle the growing volume of the data. This paper proposes a high performance design of BSP by leveraging a heterogeneous CPU/GPGPU system that consists of a host CPU and a K80 GPGPU. A series of techniques, on both data structures and execution management policies, is implemented to extensively exploit the computation capability of the heterogeneous many-core system and alleviate system bottlenecks. When compared with a parallel design on a high-end CPU, the proposed techniques achieve 48x average runtime enhancement while the maximum speedup can reach 78.76x. (C) 2018 Elsevier Inc. All rights reserved.
UNSUPERVISED HIERARCHICAL IMAGE SEGMENTATION BASED ON BAYESIAN SEQUENTIAL PARTITIONING
(2015)
Yeh, Hao-Wei
;
Tseng, Chen-Yu
;
Wu, Tung-Yu
;
Wang, Sheng-Jyh
;
電子工程學系及電子研究所
In this paper, we present an unsupervised hierarchical image segmentation algorithm based on a split-and-merge scheme. In the split phase, we propose an efficient partition algorithm, called Just-Noticeable-Difference Bayesian Sequential Partitioning (JND-BSP), to partition image pixels into a few regions, within which the color variations are perceived to be smoothly changing without apparent color differences. In the merge phase, we propose a simple but effective merging criterion to sequentially construct a hierarchical structure that represents the relative similarity among these partitioned regions. Instead of generating a segmentation result with a fixed number of segments, the new algorithm produces an entire hierarchical representation of the given image in a single run. This hierarchical representation is informative and can be very useful for subsequent processing, like object recognition and scene analysis. To demonstrate the effectiveness and efficiency of our method, we compare our new segmentation algorithm with several existing algorithms. Experiment results show that our new algorithm can not only offers a more flexible way to segment images but also provides segmented results close to human\'s visual perception.