The ACCV Workshop on Driver Drowsiness Detection from Video 2016
in conjunction with ACCV 2016, Taipei, Taiwan
Registration
Announcements
Introduction
Recent reports have suggested that drowsy driving is one of the main factors in fatal motor vehicle crashes each year. In 2014, the US National Sleep Foundation (NSF) pledged an initiative that seeks to raise public awareness on drowsy driving and asked legislators to have law enforcement, regulations and recommendations on drowsy driving and distraction prevention. Therefore, developing active monitoring systems that help drivers avoid accidents in a timely manner is of utmost importance.
Most of the previous works on drowsy driver detection focus on using limited visual cues. However, human drowsiness is a complicated mechanism. It is a challenging problem to detect driver drowsiness accurately in a timely fashion. There is lack of a publicly available video dataset to evaluate and compare different drowsy driver detection systems. This workshop will provide a great opportunity for the researchers working on the related topics, such as video event recognition or facial expression recognition, or interested in this problem to join in this competition and compare their performance with each other.
Scope
The International Workshop on Drowsy Driver Detection from Video contains two tracks: Regular Paper Track and Challenge Paper Track.
Due to the lack of a publicly available video dataset to evaluate and compare different drowsy driver detection systems, the participants of this workshop are encouraged to use the NTHU Drowsy Driver Detection (NTHU-DDD) Video Dataset and compare their performance with other participants. The top three teams with the best accuracies for the driver drowsiness detection on NTHU-DDD Video Dataset will receive award certificates and cash prizes sponsored by Qualcomm Technologies Inc.
Registration
For Challenge Paper Track, the participants should register to use the challenge dataset. For Regular Paper Track, no registration is required.
To register, please download and fill out the Registration form and Dataset License Agreement, and email them to cvlablai636@my.nthu.edu.tw.
Please note that the signed agreement form should be submitted with the registration form to complete the registration.
Please check out the website http://www.accv2016.org/call-for-proposals-for-accv2016-workshops-new for more details of ACCV Workshop on Driver Drowsiness Detection from Video 2016. We would like to invite researchers and developers to register for this Challenge and join the LinkedIn group for updated information: https://www.linkedin.com/groups/7060817. For any question, please feel free to contact the session organizers by email at (ckchiang@cs.ccu.edu.tw,lai@cs.nthu.edu.tw, msarkis@qti.qualcomm.com).
Chen-Kuo Chiang, Shang-Hong Lai, Michel Sarkis
Organizers of the ACCV Workshop on Driver Drowsiness Detection from Video 2016
Supported by: 
Introduction
Recent reports have suggested that drowsy driving is one of the main factors in fatal motor vehicle crashes each year. In 2014, the US National Sleep Foundation (NSF) pledged an initiative that seeks to raise public awareness on drowsy driving and asked legislators to have law enforcement, regulations and recommendations on drowsy driving and distraction prevention. Therefore, developing active monitoring systems that help drivers avoid accidents in a timely manner is of utmost importance.
Most of the previous works on drowsy driver detection focus on using limited visual cues. However, human drowsiness is a complicated mechanism. It is a challenging problem to detect driver drowsiness accurately in a timely fashion. There is lack of a publicly available video dataset to evaluate and compare different drowsy driver detection systems. This workshop will provide a great opportunity for the researchers working on the related topics, such as video event recognition or facial expression recognition, or interested in this problem to join in this competition and compare their performance with each other.
Scope
The International Workshop on Drowsy Driver Detection from Video contains two tracks: Regular Paper Track and Challenge Paper Track.
- Regular Paper Track: for papers related to driver drowsiness detection. The goal of this track is to identify the state-of-the-art algorithms, systems and frameworks that are particularly suitable for driver drowsiness detection.
- Challenge Paper Track: for papers participating in the challenge session of driver drowsiness detection. The challenge is held based on a driver drowsiness video dataset collected by NTHU Computer Vision Lab. The detection results of each participant will be evaluated via the training/testing videos provided.
Due to the lack of a publicly available video dataset to evaluate and compare different drowsy driver detection systems, the participants of this workshop are encouraged to use the NTHU Drowsy Driver Detection (NTHU-DDD) Video Dataset and compare their performance with other participants. The top three teams with the best accuracies for the driver drowsiness detection on NTHU-DDD Video Dataset will receive award certificates and cash prizes sponsored by Qualcomm Technologies Inc.
Registration
For Challenge Paper Track, the participants should register to use the challenge dataset. For Regular Paper Track, no registration is required.
To register, please download and fill out the Registration form and Dataset License Agreement, and email them to cvlablai636@my.nthu.edu.tw.
Please note that the signed agreement form should be submitted with the registration form to complete the registration.
Please check out the website http://www.accv2016.org/call-for-proposals-for-accv2016-workshops-new for more details of ACCV Workshop on Driver Drowsiness Detection from Video 2016. We would like to invite researchers and developers to register for this Challenge and join the LinkedIn group for updated information: https://www.linkedin.com/groups/7060817. For any question, please feel free to contact the session organizers by email at (ckchiang@cs.ccu.edu.tw,lai@cs.nthu.edu.tw, msarkis@qti.qualcomm.com).
Chen-Kuo Chiang, Shang-Hong Lai, Michel Sarkis
Organizers of the ACCV Workshop on Driver Drowsiness Detection from Video 2016
Organizers
Program Committee
Fernando De La Torre, Carnegie Mellon University
Cees Snoek, University of Amsterdam/Qualcomm Technologies Inc.
Andreas Savakis , Rochester Institute of Technology
Zhengyou Zhang, Microsoft Research
Lijun Yin, Binghamton University
Reinhard Klette, Auckland University of Technology
Gee-Sern Jison Hsu, National Taiwan University of Science and Technology
Guo-Shiang Lin, Da-Yeh University
Zhi-Fang Yang, National Taipei University
Yu-Jung Cheng, Institute for Information Industry
Ting-Lan Lin, Chung Yuan Christian University
Chang D. Yoo, Korea Advanced Institute of Science and Technology(KAIST)
Liming Chen, University of Lyon
Chun-Rong Huang,National Chung Hsing University
| Chen-Kuo Chiang, National Chung Cheng University, Taiwan |
 ckchiang@cs.ccu.edu.tw |
| Shang-Hong Lai, National Tsing Hua University, Taiwan |
| lai@cs.nthu.edu.tw |
| Michel Sarkis, Qualcomm Technologies Inc., USA |
| msarkis@qti.qualcomm.com |
Program Committee
Regular Paper Track
Challenge Paper Track
| [Aug 30, 2016] | Paper submission deadline |
| [Sep 13, 2016] | Paper decision notification |
| [Sep 18, 2016] | Camera ready paper submission |
| [Nov 24, 2016] | Workshop date |
Challenge Paper Track
| [Jul 9, 2016] | Registration to the Challenge |
| [Jul 9, 2016] | Training and evaluation datasets available to participants |
| [Aug 5, 2016] | Testing dataset released |
| [Aug 12-23, 2016] | Submission of the detection results on testing dataset |
| [Aug 25, 2016] | Accuracies for the detection results for all teams announced |
| [Aug 27, 2016] | Paper submission deadline |
| [Sep 13, 2016] | Paper decision notification |
| [Sep 18, 2016] | Camera ready paper submission |
| [Nov. 24, 2016] | Workshop date |
Evaluation Criterion
During the submission period [Aug 12-19, 2016], each team can submit up to 3 test results to adjust parameter settings in the experiments.
Each team participating in this Challenge special session need to submit their driver drowsiness detection results for the evaluation on a testing video dataset provided by the organizers. In addition, they will also need to submit a paper describing their system and performance.
The main evaluation criterion in this Challenge is the accuracy of the drowsiness detection in the testing video dataset. We will compare the submitted drowsiness detection results with the ground truth results based on the some commonly used accuracy measures, such as precision, false alarm rate, F-measure, etc.
For any queries, please email at: cvlablai636@my.nthu.edu.tw
During the submission period [Aug 12-19, 2016], each team can submit up to 3 test results to adjust parameter settings in the experiments.
Each team participating in this Challenge special session need to submit their driver drowsiness detection results for the evaluation on a testing video dataset provided by the organizers. In addition, they will also need to submit a paper describing their system and performance.
The main evaluation criterion in this Challenge is the accuracy of the drowsiness detection in the testing video dataset. We will compare the submitted drowsiness detection results with the ground truth results based on the some commonly used accuracy measures, such as precision, false alarm rate, F-measure, etc.
For any queries, please email at: cvlablai636@my.nthu.edu.tw
Detection Results Submission
Each video in the testing dataset should have one drowsiness detection result saved as a text file with the format the same as the text files for the drowsiness annotation file s for the training and evaluation datasets. For the 70 videos in the testing dataset, the submission should contain 70 detection result files placed under the same path and these text files should be compressed into a zip file.
Format of the result :
Detection File name: [video name]_drowsiness_result.txt, ex: 003_glasses_mixing_drowsiness_result.txt
Detection File format: same as the drowsiness annotation files in the training and evaluation datasets
Paper Submission Instructions
Papers should be formatted according to the ACCV 2016 Paper Kit available under paper submission on http://www.accv2016.org/paper-submission/. Papers should be submitted online by 30 August 2016 at https://cmt3.research.microsoft.com/DDDV2016.
Each video in the testing dataset should have one drowsiness detection result saved as a text file with the format the same as the text files for the drowsiness annotation file s for the training and evaluation datasets. For the 70 videos in the testing dataset, the submission should contain 70 detection result files placed under the same path and these text files should be compressed into a zip file.
Format of the result :
Detection File name: [video name]_drowsiness_result.txt, ex: 003_glasses_mixing_drowsiness_result.txt
Detection File format: same as the drowsiness annotation files in the training and evaluation datasets
Paper Submission Instructions
Papers should be formatted according to the ACCV 2016 Paper Kit available under paper submission on http://www.accv2016.org/paper-submission/. Papers should be submitted online by 30 August 2016 at https://cmt3.research.microsoft.com/DDDV2016.
Program
| 09:00-09:10 | ||
| 09:10-10:10 | ||
| 10:10-10:30 | Driver Drowsiness Detection via a Hierarchical Temporal Deep Belief Network | Jennifer Wung, Ying-Hsiu Lai, Shang-Hong Lai |
| 10:30-11:00 | ||
| 11:00-11:15 | Detection of driver drowsiness using 3D deep neural network and semi-supervised gradient boosting machine | Xuan-Phung Huynh, Sang-Min Park, Yong-Guk Kim |
| 11:15-11:30 | MSTN: Multistage Spatial-Temporal Network for Driver Drowsiness Detection | Tun-Huai Shih, Chiou-Ting Hsu |
| 11:30-11:45 | Driver drowsiness detection system based on feature representation learning using various deep networks | Sanghyuk Park, Fei Pan, Sunghun Kang, Chang D. Yoo |
| 11:45-12:00 | Representation Learning, Scene Understanding and Feature Fusion for Drowsiness Detection | Jongmin Yu , Sangwoo Park, Sangwook Lee, Moongu Jeon |
| 12:00-12:15 | Joint Shape and Local Appearance Features for Real-Time Driver Drowsiness Detection | Jie Lyu, Hui Zhang, Zejian Yuan |
| 12:15 | ||
Q & A