Monday January 29, 2018
Keynote: Image and Video Compression
Session Chair: Zoe Liu, Google, Inc. (United States)
10:40 – 11:20 AM
Sandpebble A
Dr. Yaowu Xu is currently the tech lead manager of the video coding research team at Google. The team has been responsible for developing and defining VP9, the core video technology of the WebM project. Prior to joining Google, Dr. Xu was the vice president of codec development at On2 Technologies. He was the co-creator of On2's VPx series codecs including VP32, VP4, VP5, VP6, VP7 and VP8. These codecs were broadly adopted by the industry and have fueled the phenomenal growth of web video. Dr. Xu's education includes a BS in physics, an MS and a PhD in nuclear engineering from Tsinghua University at Beijing, China. He also holds an MS and a PhD in electrical and computer engineering from the University of Rochester. Dr. Xu has published many technical papers in the area of image processing on leading journals and international conferences. He also holds many patents and has numerous patent applications pending in the area of digital video compression. Dr. Xu's research and development experiences include digital video compression and processing, real time video encoding and decoding, mobile video, image processing, pattern recognition and machine learning. His current research focuses on advanced algorithms for digital video compression.
VIPC-123
Technical overview of AV1: An open source video codec from the Alliance for Open Media, Yaowu Xu, Google, Inc. (United States)
Image and Video Compression
Session Chair:
Zoe Liu, Google, Inc. (United States)
11:20 AM – 12:40 PM
Sandpebble A
11:20VIPC-153
CrossEncoders: A complex neural network compression framework, Chirag Agarwal, Mehdi Sharifzadeh, and Dan Schonfeld, University of Illinois at Chicago (United States)
11:40VIPC-154
Multi-level machine learning-based early termination in VP9 partition search, Yang Xian1, Yunqing Wang2, Yingli Tian1, Yaowu Xu2, and James Bankoski2; 1The City University of New York and 2Google Inc. (United States)
12:00VIPC-155
Texture segmentation based video compression using convolutional neural networks, Chichen Fu1, Di Chen1, Edward Delp1, Zoe Liu2, and Fengqing Zhu1; 1Purdue University and 2Google, Inc. (United States)
12:20VIPC-156
Multi-reference video coding using stillness detection, Di Chen1, Zoe Liu2, Yaowu Xu2, Fengqing Zhu1, and Edward Delp1; 1Purdue University and 2Google, Inc. (United States)
12:40 – 2:00 PM Lunch
2:00 – 3:00 PM PLENARY: Overview of Modern Machine Learning and Deep Neural Networks - Impact on Imaging and the Field of Computer Vision
3:00 – 3:30 PM Coffee Break
Deep Learning for Image Analysis
Session Chair:
Fengqing Zhu, Purdue University (United States)
3:30 – 4:50 PM
Sandpebble A
3:30VIPC-173
Event recognition in personal photo collections: An active learning approach, Kashif Ahmad, Mohamed Mekhalfi, and Nicola Conci, Università degli Studi di Trento (Italy)
3:50VIPC-174
Generative adversarial networks for open set historical Chinese character recognition, Xiaoyi Yu, Jun Sun, and Satoshi Naoi, Fujitsu R&D Co. Limited (China)
4:10VIPC-175
Using convolutional neural networks and transfer learning for Content-Based Image Retrieval (CBIR), Felix Mayer, Maximilian Li, Martin Steinebach, and Marcel Schaefer, Fraunhofer SIT (Germany)
4:30VIPC-176
Approach for machine-printed Arabic character recognition: The-state-of-the-art deep-learning method, Daegun Ko, Changhyung Lee, Donghyeong Han, Hyeongsu Ohk, Ki-Min Kang, and Seong-Wook Han, HP Inc. (Republic of Korea)
5:00 – 6:00 PM All-Conference Welcome Reception
Tuesday January 30, 2018
7:15 – 8:45 AM Women in Electronic Imaging Breakfast
Computer Vision Applications in Sports
Session Chair:
Edward Delp, Purdue University (United States)
9:10 – 9:50 AM
Sandpebble A
9:10VIPC-205
Toward automatic and objective evaluation of synchronization in video of synchronized diving, Xin Li and Yixin Du, West Virginia University (United States)
9:30VIPC-206
Convolutional neural networks for the analysis of broadcasted tennis games, Grigorios Tsagkatakis1, Mustafa Jaber2, and Panagiotis Tsakalides1; 1Foundation for Research and Technology (FORTH) (Greece) and 2NantVision Inc. (United States)
10:00 AM – 7:30 PM Industry Exhibition
9:50 – 10:40 AM Coffee Break
Keynote: Image and Video Analytics
Session Chair: Grigorios Tsagkatakis, Foundation for Research and Technology (FORTH) (Greece)
10:40 – 11:20 AM
Sandpebble A
Dr. Ioannis Katsavounidis received the Diploma (BS/MS) from the Aristotle University of Thessaloniki, Greece, (1991) and his MS and PhD from the University of Southern California, Los Angeles, (1992 and 1998 respectively), all in electrical engineering. From 1996 to 2000, he worked in Italy as an engineer for the high-energy physics department of the California Institute of Technology. From 2000 to 2007, he worked at InterVideo, Inc., in Fremont, CA, as director of software for advanced technologies, in charge of MPEG2, MPEG4 and H.264 video codec development. Between 2007 and 2008, he served as CTO of Cidana, a mobile multimedia software company in Shanghai, China, covering all aspects of DTV standards and codecs. From 2008 to 2015 he was an associate professor with the department of electrical and computer engineering at the University of Thessaly in Volos, Greece, teaching undergraduate and graduate courses in signals, controls, image processing, video compression, and information theory. He is currently a senior research scientist at Netflix, working on video quality and video codec optimization problems. His research interests include image and video quality, compression and processing, information theory, and software-hardware optimization of multimedia applications.
VIPC-215
Perceptual optimization in video coding - a systematic approach, Ioannis Katsavounidis, Netflix (United States)
Image and Video Analytics
Session Chair:
Grigorios Tsagkatakis, Foundation for Research and Technology (FORTH) (Greece)
11:20 AM – 12:40 PM
Sandpebble A
11:20VIPC-251
Interactive hand pose estimation: Boosting accuracy in localizing extended finger joints, Cairong Zhang, Guijin Wang, Hengkai Guo, Xinghao Chen, Fei Qiao, and Huazhong Yang, Tsinghua University (China)
11:40VIPC-252
Efficient preprocessing and feature extraction for robust face recognition, Huda Algharib and Osman Serdar Gedik, Ankara Yildirim Beyazit (Turkey)
12:00VIPC-253
Text/figure separation in document images using Docstrum descriptor and two-level clustering, Valery Anisimovskiy, Ilya Kurilin, Andrey Shcherbinin, and Petr Pohl, Samsung R&D Institute Russia (Russian Federation)
12:20VIPC-254
Prediction system for activity recognition with compressed video, Chengzhang Zhong and Amy Reibman, Purdue University (United States)
12:40 – 2:00 PM Lunch
2:00 – 3:00 PM PLENARY: Fast, Automated 3D Modeling of Buildings and Other GPS Denied Environments
3:00 – 3:30 PM Coffee Break
Imaging Systems
Session Chair:
Amy Reibman, Purdue University (United States)
3:30 – 4:30 PM
Sandpebble A
3:30VIPC-263
A robust and accurate calibration method for out-of-focus camera, Xiaowei Hu1, Guijin Wang1, Jinnan Wang1, Pengfei Sun1, Jingtao Fan1, Feng Chen1, and Yiyuan Xie2; 1Tsinghua University and 2Southwest University of Science and Technology (China)
3:50VIPC-264
Improving the efficiency of on-site operators in utility management: Combining hololens and AR for real-time check of electricity meters, Nicola Conci1, Lorenzo Orlandi1,2, and Daniele Sevegnani2; 1Università degli Studi di Trento and 2ARCODA (Italy)
4:10VIPC-265
Generation of stereoscopic image sequences from monocular videos using epipolar geometry, Vasundhara Goyal and Dan Schonfeld, University of Illinois at Chicago (United States)
5:30 – 7:30 PM EI 2018 Symposium Demonstration Session
Wednesday January 31, 2018
Visual Information Processing and Communication IX Interactive (Poster) Papers Session
5:30 – 7:00 PM
The Grove
The following works will be presented at the EI 2018 Symposium Interactive Papers Session.
VIPC-414
An interrupted projection using seam carving for 360-degree images, Ikuko Tsubaki and Kazuo Sasaki, Tokyo University of Technology (Japan)
5:30 – 7:00 PM Meet the Future: A Showcase of Student and Young Professionals Research