電腦視覺與虛擬實境組
Computer Vision and Virtual Reality Group

電腦視覺組致力於智慧型運輸系統的開發,主要的研究方向為駕駛警示的輔助系統,
包含行人偵測、車輛偵測等主題,營造未來友善的交通環境。
The Computer Vision Group dedicates to the development of Intelligent Transportation System.
The assistive alarming system to drive is our major research area, including pedestrian detection and vehicle detection.
虛擬實境組的研究專注於個人化的動作辨識,將辨識的技術使用於人機互動、擴增實境等應用,
能在未來以虛擬實境輔助人類日常活動及提升生活的便利性。
The Virtual Reality Group focus on the personalized action recognition
which is applied in Human-Machine Interaction and Augmented Reality.
The technique is expected to assist human’s daily activity as well as to enhance the convenience of future life.

Research

智慧型運輸系統

智慧型運輸系統
Intelligent Transportation System (ITS)

在智慧型型運輸系統中,先進駕駛補助系統(ADAS)是一個重要的議題:經由車載的處理器,應用攝影機或與其他感應技術的結合獲得車輛前方道路場景的資訊,並分析或指出是否存在危險的情況。這種系統提供一個警示的裝置避免駕駛因疲勞或短暫的不注意而發生車禍。經由上述的理由,我們發展出一連串以電腦視覺(computer vision)為基礎的技術,包含了道路線辨識和障礙物偵測(行人或車輛偵測)。

Advanced Driver Assistance System (ADAS) is an important topic in Intelligent Transportation Systems (ITS), which applies the cameras or other sensor fusion techniques to obtain information from road scene in front of vehicle, and prevents from the dangerous situations. The systems in the topic provide a warning mechanism to protect drivers from car accidents caused by fatigue or sudden distraction. According to this reason, we have developed a series of techniques based on computer vision techniques including lane recognition and obstacle detection (such as pedestrian or vehicle detection).

道路障礙物偵測

道路障礙物偵測
On-road obstacle Detection

道路障礙物偵測可以說是先進駕駛補助系統的第一關。為了在複雜的行車道路環境中偵測出多種類的道路障礙物,我們使用了卷積神經網路作為偵測的骨幹。進一步提出了新型的偵測架構並發展創新之深度學習框架降低計算複雜度以應用於車載嵌入式系統。

The motivation of system is that many car accidents happened due to collisions with on-road obstacle. Here we have developed on-road obstacle detection systems base on convolutional neural network(CNN). This system is a proposal-free framework, so it is fast and efficient. And it has been widely applied to Intelligence Transportation Systems or some applications of surveillance.

基於深度學習語義分割之城市道路汽車轉向操控

基於深度學習語義分割之城市道路汽車轉向操控
A Deep Learning Based Semantic Segmentation Approach for Car Steering on Urban Roads

在視覺式自動駕駛系統中,感知與控制是兩個重要且待解決的議題。我們提出了一個使用語義感知並基於端對端深度卷積神經網路的方法來解決自動駕駛中的視覺式控制問題。在第一個階段中,使用一個深度卷積神經網路從輸入影像產生語義分割的結果,在第二個階段中則使用另一個深度卷積神經網路從語義分割資訊來預測出汽車轉向操控。證明了語義分割可以用來提升視覺式自動駕駛系統的效能。

In vision based autonomous driving systems, perception and control tasks are two critical problems to be solved. We propose an end-to-end CNN architecture with semantic perception to solve the vision based control problem in autonomous driving. In the first stage, a CNN module is used to generate semantic segmentation from the input image. In the second stage, another CNN module is used to take advantage of the semantic perception to predict steering controls. We show that semantic segmentation can be applied to enhance the performance of a vision based autonomous driving system.

個人化的動作辨識

個人化的動作辨識
User-Specific Motion Recognition

個人化動作辨識研究之目的在於想要建立一個可靠的通用型動作辨識系統,其中之困難處在於適應不同使用者有不同的動作習慣與外觀,針對這些難題尋求解決之道,則將可以迅速有效地建立個人化的動作辨識系統,進而運用在即時的互動應用,能以動作為輸入訊號融合於仰賴固定型態動作辨識的系統,提升未來日常生活上的便利性以及在虛擬實境的延伸技術上有多方面的發展。

This research focuses on developing a reliable and general action-recognition (AR) system, which is difficult to adapt various appearance and action styles of different users. Many future applications rely on this technique such as the real-time Human-Machine Interaction (HCI) systems due to the significance of identifying the input signal of a certain motion pattern, and are potentially used in extension techniques of AR as well as enhance the convenience of life.

Members

TitleNameDurationEmail
博士研究生 Ph.D. student湯雅惠 Ya-Hui Tang(2017-)tang.aggie@gmail.com
碩士研究生 M.S. student丁柏文 Po-Wen Ting(2016-)ck980046@gmail.com
碩士研究生 M.S. student葉興宇 Hsing-Yu Yeh(2016-)yehhsingyu1029@qq.com
碩士研究生 M.S. student韓翔宇 Hsiang-Yu Han(2016-)andy8517251@gmail.com
碩士研究生 M.S. student周恩德 En-Te Chou(2017-)csjou88@gmail.com
碩士研究生 M.S. student陳禹齊 Yu-Ci Chen(2017-)yucharlie.chen@gmail.com
碩士研究生 M.S. student洪子翔 Zih-Siang Hong(2017-)asas880015@gmail.com
畢業學生沈宗穎 Zong-Ying Shen(2015-2017)
畢業學生吳忞諭 Min-Yu Wu(2015-2017)
畢業學生塗國星 Kuo-Hsin Tu(2015-2017)
畢業學生謝宇勳 Joseph Shie(2014-2016)
畢業學生陳子揚 Zi-yang Chen(2014-2016)
畢業學生詹益銘 Yi-Ming Chan(2005-2016)
畢業學生林君丞 Chun-Cheng Lin(2013-2015)
畢業學生吳惟郁 Wei-Yu Wu(2013-2015)
畢業學生柳成荫 Cheng-yin Liu(2012-2015)
畢業學生陳冠伃 Guan-yu Chen(2013-2015)
畢業學生陳涵軒 Han-Hsuan Chen(2012-2014)
畢業學生黃邦庭 Pang-Ting Huang(2012-2014)
畢業學生許彥彬 Yan-Bin Xu(2012-2014)
畢業學生吳承恩 Cheng-En Wu(2011-2013)
畢業學生陳逸祥 Yi-Hsiang Chen(2011-2013)
畢業學生蔡日昇 Ri-Sheng Cai(2011-2013)
畢業學生符劼 Jie Fu(2011-2013)
畢業學生高毓甫 Yu-Fu Kao(2010-2012)
畢業學生李依書 Yi-Shu Lee(2010-2012)
畢業學生黃恩暐 En-Wei Huang(2002-2012)
畢業學生翁藝睿 Yi-Rui Weng(2010-2012)
畢業學生莊理安 Li-An Chuang(2009-2011)
畢業學生莊珞傑 Luo-Chieh Chuang(2009-2011)
畢業學生林預淳 Yu-Chun Lin(2008-2010)
畢業學生林斌峰 Bin-Feng Lin(2008-2010)
畢業學生鄭一中 Yi-Zhong Zheng(2009-2010)
畢業學生許平昇 Ping-Sheng Xu(2008-2010)
畢業學生李忞藯 Min-Wei Li(2007-2009)
畢業學生林志鴻 Zhi-Hong Lin(2007-2009)
畢業學生洪思穎 Ssu-Ying Hung(2006-2008)
畢業學生莊振勳 Cheng-Hsiung Chuan(2006-2008)
畢業學生邱一航 Yi-Hang Chiu(2006-2008)
畢業學生楊嘉鳴 Jia-Ming Yang(2006-2008)
畢業學生袁維均 Wei-Chun Yuan(2005-2007)
畢業學生黃贊宇 Chan-Yu Huang(2005-2007)
畢業學生黃世勳 Shih-Shinh Huang(1999-2007)
畢業學生王劭廷 Shao-Ting Wang(2005-2007)
畢業學生陳任志 Jerry Chen(2004-2006)
畢業學生蔡濬帆 Jyun-Fan Tsai(2004-2006)
畢業學生黃俊棋 Jun-Qi Huang(2004-2006)
畢業學生王俊哲 Chun-Che Wang(2002-2004)
畢業學生陳琮仁 Chun-Ren Chen(2001-2003)
畢業學生謝衛中 Wei-Chung Hsieh(2000-2002)

Journal Paper

Title
Online view-invariant human action recognition using rgb-d spatiotemporal matrix Yen-Pin Hsu, Chengyin Liu, Tzu-Yang Chen, Li-Chen Fu Pattern Recognition (PR), vol. 60, pp.215-226, Dec 2016
Pang-Ting Huang; Yi-Ming Chan; Li-Chen Fu; Shih-Shinh Huang; Pei-Yung Hsiao; Wei-Yu Wu; Chun-Cheng Lin; Kuo-Ching Chang; Ping-Min Hsu, "Pedestrian detection system in low illumination conditions through Fusion of image and range data," Intelligent Transportation Systems (ITSC), 2014 IEEE 17th International Conference on , vol., no., pp.2253,2254, 8-11 Oct. 2014
Han-Hsuan Chen; Chun-Cheng Lin; Wei-Yu Wu; Yi-Ming Chan; Li-Chen Fu; Pei-Yung Hsiao, "Integrating appearance and edge features for on-road bicycle and motorcycle detection in the nighttime," Intelligent Transportation Systems (ITSC), 2014 IEEE 17th International Conference on , vol., no., pp.354,359, 8-11 Oct. 2014
Yi-Ming Chan, Shih-Shinh Huang, Li-Chen Fu, Pei-Yung Hsiao, and Min-Fang Lo, “Vehicle Detection and Tracking under Various Lighting Conditions Using Particle Filter,” IET Transactions on Intelligent Transportation Systems, 2012.
Bin-Feng Lin, Yi-Ming Chan, Li-Chen Fu, Pei-Yung Hsiao, Li-An Chuang, Shin-Shinh Huang, and Min-Fang Lo, “Integrating Appearance and Edge Features for Sedan Vehicle Detection in the Blind-spot Area,” IEEE Transactions on Intelligent Transportation Systems, 2012.
Shih-Shinh Huang, Li-Chen Fu, and Pei-Yung Hsiao, 「Region-Level Motion-Based Background Modeling and Subtraction Using MRFs,」 IEEE Trans. on Image Processing (SCI) Vol. 16, No. 5, pp. 1446-1456, May 2007.
Shih-Shinh Huang, Li-Chen Fu, and Pei-Yung Hsiao, 「Region-Level Motion-Based Foreground Segmentation under a Bayesian Network,」 Submitted to IEEE Trans. on Circuits and Systems for Video Technology (SCI), Feb. 2007.
Shih-Shinh Huang, Li-Chen Fu, Pei-Yung Hsiao, 「Region-Level Motion-Based Background Modeling and Subtraction Using MRFs,」 Lecture Notes in Computer Science, 2006.
Yi-Ming Chan, Shih-Shinh Huang, Li-Chen Fu, and Pei-Yung Hsiao, 「Vehicle Detection under Various Lighting Conditions by Incorporate Particle Filter,」 Journal of Vehicle Engineering, Vol.3, pp.31-44 , May 2006.
Jyun-Fan Tsai, Shih-Shinh Huang, Chan-Yu Huang, Li-Chen Fu, and Pei-Yung Hsiao, 「On Road Image Acquiring and Anti-Blooming System at Nighttime by Using High Dynamic Image Reconstruction with Motion Compensation,」 Journal of Vehicle Engineering, Vol.3, pp.45-58, May 2006.
Pei-Yung Hsiao, Hsien-Chein Cheng, Shih-Shinh Huang, and Li-Chen Fu, 「A Mixed Signal CMOS Imager with Lane Detector for Use in Smart Vehicles,」 Submitted to IEEE Trans. on Vehicular Technology (SCI), Oct. 2006.
Shih-Shinh Huang, Li-Chen Fu, Pei-Yung Hsiao, 「Silhouette-Based Human Pose Estimation Using Reversible Jump Markov Chain Monte Carlo,」 IEE Electronics Letters, 2006.
Chun-Che Wang, Shih-Shinh Huang, Pei-Yung Hsiao, and Li-Chen Fu, 「Computer Vision-Based Lane Detection and Vehicle Recognition for Driver Assistance System at Night,」 Journal of Vehicle Engineering, Vol.2, pp.51-64 , May 2005.

Conference Paper

Title
Robust Dynamic Hand Gesture Recognition System with Sparse Steric Haar-Like Feature for Human Robot Interaction Chengyin Liu, Tzu-Yang Chen, Li-Chen Fu SICE Annual Conference, 2016
Daily Activity Recognition Using the Informative Features from Skeletal and Depth Data Min-Yu Wu, Tzu-Yang Chen, Kuan-Yu Chen, Li-Chen Fu IEEE International Conference on Robotics and Automation, 2016
Deep Learning for Integrated Hand Detection and Pose Estimation Tzu-Yang Chen, Min-Yu Wu, Yu-Hsun Hsieh, Li-Chen Fu International Conference on Pattern Recognition, 2016
Cheng-En Wu, Yi-Ming Chan, Li-Chen Fu, Pei-Yung Hsiao, Shin-Shinh Huang, Han-Hsuan Chen, Pang-Ting Huang and Shao-Chung Hu, "Combining Multiple Complementary Features for Pedestrian and Motorbike Detection," International IEEE Conference on Intelligent Transportation Systems, pp. 1358-1363, 2013
Yi-Ming Chan, Li-Chen Fu, Pei-Yung Hsiao, and Min-Fang Lo, "Pedestrian Detection Using Histograms of Oriented Gradients of Granule Feature," IEEE Intelligent Vehicle Symposium, pp. 1410-1415, 2013.
Yi-Shu Lee, Yi-Ming Chan, Li-Chen Fu, Fellow, IEEE, Pei-Yung Hsiao, Li-An Chuangs, Yi-Hsiang Chen and Ming-Fang Luo, "Nighttime Pedestrian Detection by Selecting Strong Near-Infrared Parts and Enhanced Spatially Local Model," International IEEE Conference on Intelligent Transportation Systems, pp. 1765-1770, 2012.
Yu-Fu Kao, Yi-Ming Chan, Li-Chen Fu, Pei-Yung Hsiao, "Comparison of Granules Features for Pedestrian Detection," International IEEE Conference on Intelligent Transportation Systems, pp. 1777-1782, 2012.
Yu-Chun Lin, Yi-Ming Chan, Luo-Chieh Chuang, Li-Chen Fu, Shih-Shinh Huang, Pei-Yung Hsiao, and Min-Fang Luo, “Near-Infrared Based Nighttime Pedestrian Detection by Combining Multiple Features,” Proceedings of IEEE International Conference on Intelligent Transportation Systems, pp. 1549-1554, Washington, DC, 2011.
Bin-Feng Lin, Yi-Ming Chan, Li-An Chuang, Li-Chen Fu, Pei-Yung Hsiao, and Shih-Shinh Huang, 「Incorporating Appearance and Edge Features for Vehicle Detection in the Blind-Spot Area,」 13th International IEEE Conference on Intelligent Transportation Systems, Madeira Island, Portugal, 2010.
Ssu-Ying Hung, Yi-Ming Chan, Bin-Feng Lin, Li-Chen Fu, Pei-Yung Hsiao, Shin-Shinh Huang, 「Tracking and detection of lane and vehicle integrating lane and vehicle information using PDAF tracking model,」 12th International IEEE Conference on Intelligent Transportation Systems, pp.603-608, 2009
Cheng-Hsiung Chuang, Shih-Shinh Huang, Li-Chen Fu and Pei-Yung Hsaio, 「Monocular Multi-Human Detection Using Augmented Histograms of Oriented Gradients」, IEEE International Conference on Pattern Recognition, 2008.
Chan-Yu Huang, Shih-Shinh Huang, Yi-Ming Chan, Yi-Hang Chiu, Li-Chen Fu and Pei-Yung Hsaio, 「Driver Assistance System Using Integrated Information from Lane Geometry and Vehicle Direction」, IEEE International Conference on Intelligent Transportation Systems, 2007.
Yi-Ming Chan, Shih-Shinh Huang, Li-Chen Fu and Pei-Yung Hsiao, 「Vehicle Detection under Various Lighting Conditions by Incorporating Particle Filter」, IEEE International Conference on Intelligent Transportation Systems, 2007.
Shih-Shinh Huang, Li-Chen Fu, Pei-Yung Hsiao, 「A Bayesian Framework for Foreground Segmentation」, IEEE International Conference on Systems, Man, and Cybernetics, 2006.
Shih-Shinh Huang, Li-Chen Fu, Pei-Yung Hsiao, 「Silhouette-Based Human Pose Estimation Using Reversible Jump Markov Chain Monte Carlo,」 IEEE International Conference on Pattern Recognition, 2006.
Shih-Shinh Huang, Li-Chen Fu, Pei-Yung Hsiao, 「A Framework for Human Pose Estimation by Integrating DD-MCMC and MOEA,」 IEEE International Conference on Robotics and Automation, 2006.
Shih-Shinh Huang, Li-Chen Fu, Pei-Yung Hsiao, 「Region-Level Motion-Based Foreground Detection with Shadow Removal Using MRFs,」 Asian Conference on Computer Vision, 2006.
Jyun-Fan Tsai, Shih-Shinh Huang, Yi-Ming Chan, Chan-Yu Huang, Li-Chen Fu, and Pei-Yung Hsiao, 「Road Detection and Classification in Urban Environments Using Conditional Random Field Models,」 IEEE International Conference on Intelligent Transportation System, 2006.
Shih-Shinh Huang, Li-Chen Fu, Pei-Yung Hsiao, 「A Region-Level Motion-Based Background Subtraction Using MRFs,」 IEEE International Conference on Robotics and Automation, 2005.
Chun-Che Wang, Shih-Shinh Huang, Li-Chen Fu, and Pei-Yung Hsiao, 「Driver Assistance System for Lane Detection and Vehicle Recognition with Night Vision,」 IEEE International Conference on Intelligent Robots and Systems, 2005.
Shih-Shinh Huang, Li-Chen Fu, Pei-Yung Hsiao, 「A Region-Based Background Modeling and Subtraction Using Partial Directed Hausdorff Distance,」 IEEE International Conference on Robotics and Automation, 2004.
Shih-Shinh Huang, Chung-Jen Chen, Pei-Yung Hsiao, and Li-Chen Fu, 「On-Board Vision System for Lane Recognition and Front-Vehicle Detection to Enhance Driver's Awareness,」 IEEE International Conference on Robotics and Automation, 2004.
W. C. Hsieh, Li-Chen Fu and Shih-Shinh Huang, 「Vision Based Obstacle Warning System for On-Road Driving,」 IEEE Conference on Intelligent Robots and Systems, 2003.
Z. T. Sun, Li-Chen Fu and Shih-Shinh Huang, 「On-Road Computer Vision Based Obstacle Detection,」 IEEE Conference on Intelligent Robots and Systems, 2002.
C. Y. Liu, Li-Chen Fu, Shih-Shinh Huang, 「Computer Vision Based Objects Detection and Recognition for Safe Vehicle Driving,」 IEEE Conference on Robotics and Automation, 2001.