News

We are so happy to welcome a senior undergraduate student –Chuanzi He to our lab in 2016 summer. Let us hear what she said about the intern in our group.

1. Could you briefly introduce yourself? (Previous research experience, project experience, research interest and expertise)
I am a senior undergraduate from Fudan University, Shanghai. My research experience starts with image processing — single image haze removal algorithm, which is to restore haze-free images from hazy ones. In order to solve one segmentation problem in dehazing, I went to learn more about computer vision and machine learning techniques, which lately became my research interest. They are just so fascinating.

2.What’s your first impression of USC and MCL?

Encountered a fluffy squirrel near EEB the first day I came! This is a lively campus with beautiful architecture. MCL people are nice, insightful and helpful. Really appreciate those carefully-prepared summer courses and all the help they offered. Other summer interns are also helpful and diligent. It is so great that I could spend a whole summer with this group of bright individuals.

3. What’s your future expectation for MCL?

I hope to gain more insights in deep learning architectures and to strengthen my background in visual recognition.

Yuzhuo Ren attended CVPR 2016, hosted in Las Vegas from June 26th to July 1st, to present her work in the Large-scale Scene Understanding Challenge (LSUN) workshop. The website of the LSUN challenge is provided here: http://lsun.cs.princeton.edu/2016/.

There are three challenges or tasks in the LSUN workshop: scene classification, saliency predication and room layout estimation. The workshop provides datasets that are much larger than existing ones for benchmarking algorithms proposed by different teams. The team with the best performance is named the winner and is invited to give a talk during the workshop.

Yuzhuo presented her work during the LSUN workshop. The title of her work is “A Coarse-to-Fine Indoor Layout Estimation (CFILE) Method.” This method provides a two-stage indoor layout estimation system, coarse layout estimation followed by layout refinement. The talk was very well received.

The two keynote speakers, Jitendra Malik and Yann LeCun, gave very insightful talks. Jitendra Malik discussed semantic segmentation using RGBD data, while Yann LeCun’s talk was focused on object detection.

In the award session, Jitendra Malik, Yann LeCun, the LSUN workshop organizers and challenge winners had a photo taken together.

Professor Jay Kuo has joined an international team led by Professor Ed Delp of Purdue University to develop a system for military intelligence that would detect doctored images and video and determine specifically how they were manipulated.

A huge volume of images and video of potential intelligence value are uploaded daily to the Internet. However, visual media are easily manipulated using software tools that are readily available to the public. The researchers will strive to create an “end-to-end” system capable of handling the massive volume of media uploaded regularly to the Internet. Such a system also could have potential commercial applications, representing a potential tool for news and social media platforms to authenticate images and video before posting them.

The project is funded by the U.S. Defense Advanced Research Projects Agency. The team’s co-principal investigators are Walter Scheirer, Kevin W. Bowyer, and Patrick J. Flynn from the University of Notre Dame; Anderson Rocha from the University of Campinas in Brazil; C.-C. Jay Kuo from the University of Southern California; Paolo Bestagini and Stefano Tubaro at Politecnico di Milano in Italy; Mauro Barni at the University of Siena in Italy; and Nasir Memon at New York University.

Professor C.-C. Jay Kuo gave a keynote speech at the 8th International Conference on Quality of Multimedia Experience (QoMEX) in Lisbon, Portugal, on June 6th. The title of his talk is “Perceptual Coding: Hype or Hope?”. The abstract of his talk is given below.
“There has been a significant progress in image/video coding in the last 50 years, and many visual coding standards have been established, including JPEG, MPEG-1, MPEG-2, H.264/AVC and H.265, in the last three decades. The visual coding research field has reached a mature stage, and the question “is there anything left for image/video coding?” arises in recent years. One emerging R&D topic is “perceptual coding”. That is, we may leverage the characteristics of the human visual system (HVS) to achieve a higher coding gain. For example, we may change the traditional quality/distortion measure (i.e., PSNR/MSE) to a new perceptual quality/distortion measure and take visual saliency and spatial-temporal masking effects into account. Recent developments in this area will be reviewed first. However, “is it sufficient to keep visual coding research vibrant and prosperous for another decade with such a modification?” The answer is probably not. In this talk, I will present a new HVS-centric coding framework that is dramatically differently from the past. This framework is centered on two key concepts – the stair quality function (SQF) and the Just-Noticeable-Differences (JND). It will lead to numerous new R&D opportunities and revolutionize coding research with modern machine learning tools.”
Perceptual coding is one of the main activities of the MCL in last 5-6 years. Several PhD students and post-doc have made contributions to this topic, including Sudeng Hu, Yu-Chieh Lin, Lina Jin and Haiqiang Wang. This project has been supported by Netflix [...]

Seven MCL members attended the Viterbi PhD hooding ceremony on Thursday, May 12, 2016, from 8:30-11:00 a.m. in the Bovard Auditorium. They were Young Ju Jeong, Xiaqing Pan, Chen Chen, Shangwen Li, Jian Li, Chun-Ting Huang  and Hao Xu. Congratulations to them and their families for their accomplishments in completing their PhD program at USC.

Ms. Young Ju Jeong joined the MCL in 2005 Fall firstly as a MS student and then a PhD student. Young Ju took a leave of absence, returned to Korea and worked as a research engineer in the Samsung Advanced Institute of Technology (SAIT) in 2007 August. She has resumed her PhD program remotely since two years ago by taking courses through DEN and communicating with Professor Kuo on her research progress.Her dissertation title is “Autostereoscopic 3D Display Rendering from Stereo Sequences”.

Xiaqing Pan received his B.S. degree in University of Shanghai for Science and Technology (USST), Shanghai, China in 2010 and M.S.E.E. degree in USC, Los Angeles, U.S.A. in 2012. He joined MCL in 2012 Fall. His research interests include 3-D mesh classification and retrieval, super-resolution and computer vision.

Chen Chen received his bachelor degree from the Beijing University of Posts and Telecoms and earned his M.S. at the University of Southern California. In Prof. Kuo’s Media and Creative Lab, he focuses on computer vision research topics.

Jian Li received his B.S. degree from department of Electronic Science and Technology at University of Science and Technology of China (USTC) in 2009. He obtained his M.S. degree in Electrical Engineering from University of Southern California (USC) in 2011. He is currently pursuing PhD degree in Electrical Engineering at USC. He is a Research Assistant at Media Communications Lab advised by Prof. C.-C. Jay Kuo. His [...]