News

Ping Wang is a MS student majoring in Electrical Engineering (Multimedia) at USC. She is conducting directed research (DR) at MCL since the beginning of Fall 2015. We had an interview with her, talking about her research at MCL.

1. Could you please briefly introduce your research at MCL?

My research focuses on quality assessment of compressed image/video based on human’s visual perception. In this research, my group developed a Just Noticeable Difference (JND)-based image/video quality assessment metric, and conducted JND subjective test to collect data. A method to process the measured raw data is proposed afterwards, from which a human-centric quality database is constructed.

2.  Why did you choose MCL to conduct your DR?

It has to do with my major. I’m a multimedia-focusing MS student who has great passion in image/video processing and related technology. I have been taken courses in the previous three semesters yet haven’t got a taste of researching. I feel that doing a research in my cherished field is a perfect end mark for my master’s degree.

3. It’s coming to your end of DR. How would you evaluate it? Do you find it helpful for your study or future career at USC? What do you gain from this experience?

I think this short DR period at MCL is a very unforgettable experience in my whole life. Here I was provided tons of resources. It’s an amazing group where I can always find new things to learn, where each member is helpful and willing to share their experience, where everybody including Professor Kuo is working closely together and where other people’s wisdom can actually inspire me a lot. I’m pretty happy that my master’s degree is fulfilled by the valuable experience here in MCL.

Sudeng Hu, a MCL member, has passed his defense on Oct 26, 2015. Congratulations!

His dissertation title is “Techniques for Compressed visual data quality assessment and advanced video coding”. Object quality assessment for compressed images and videos is critical to various image and video compression systems that are essential in the delivery and storage. In the thesis, an image quality metric (IQM) and a video quality metric (VQM) are proposed based on perceptually weighted distortion in term of the MSE. To capture the characteristics of HVS, for images, a spatial randomness map is proposed to measure the masking effect and a preprocessing scheme is proposed to simulate the processing that occurs in the initial part of human HVS. For the VQM, the dynamic linear system is employed to model the video signal and is used to capture the temporal randomness of the videos. The performance of the proposed IQM and VQM are validated on various image and video databases with various compression distortions. The experimental results show that the proposed IQM and VQM outperforms other benchmark quality metrics.

Sudeng gave a nice talk with clarity and smooth flow. The Committee was impressed by his high quality research work and results. When talking about his success in his research work, Sudeng shared his experience with us. He believes that 4 years of PhD life gives him many good memories to have in the rest of his life. He has been enjoying working and studying with our group members since the first day he joined the MCL lab. He also enjoyed conducting research here under the guidance of Prof. Kuo. Prof. Kuo and him had an interesting research topic with great challenges. They had hard time to conquer [...]

Dr. Jonghye Woo, an MCL alumnus, was recently promoted to Assistant Professor at Harvard Medical School. Congratulations!

Dr. Woo graduated from MCL in 2009. He became a research associate at Cedars-Sinai Medical Center in 2010 and joined University of Maryland and John Hopkins University as post-doc in 2012. After that he became a faculty member at MGH/Harvard. His research interests lie in medical imaging, particularly multimodal MRI analysis of speech, cardiac/tongue motion modeling and analysis, etc.

Talking about his career path, he thinks it was deeply influenced by his PhD journey, during which he was advised by Prof. Kuo and met several well-known people in the field of medical imaging. Recalling the days in MCL, he felt thankful that he joined the large MCL group, where he interacted a lot with his fellows from various sub-groups. Those experiences are helpful because communication skills are crucial to start career in academia.

Dr. Woo also gave several suggestions to those who want to pursue success in academia. To become a researcher, one needs to be highly motivated and always keeps a clear goal. Humbleness is another important characteristic, which makes one open to different ideas and collaboration with other researchers. The last characteristic he emphasized is persistence. Research is challenging and persistence helps one endure the inevitable struggling.

MCL has a new PhD student, Ye Wang, in Fall 2015. Let’s give him a warm welcome!

Ye received M.S. degree from Peking University and B.S. degree from Xi’an Jiaotong University in 2014 and 2011, respectively. He joined MCL to pursue his PhD degree in Fall 2015. We had a briefly interview with him.

1. Could you briefly introduce yourself? (Previous research experience, project experience, research interest and expertise)

My previous research field focuses on semiconductor fabrication and test, especially in GaN enhancement mode MOSFET. In 2012 I had a wonderful opportunity to participate in a 3-year project in ‘National Science and Technology Major Project 02’. In the meantime, I published two IEEE journal papers and three top conference papers as well as an invited conference paper.

Since arriving at USC, I have heard more and more about computer vision and machine learning, which are always my keen interest that I haven’t got chances to learn before. Therefore I really appreciate Prof. Kuo for giving me this opportunity to study in MCL and explore more in this exciting area.

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

USC has a beautiful campus, and Trojans are like a big family with great diversity. Since MCL is a very large group, I was curious about how Prof. Kuo could guide so many students.

However, when I became a member of this group, I finally understand the way Prof. Kuo inspires the students and how they all work together closely and efficiently. Moreover, their passion for research, self-discipline and sound knowledge in this field make me feel really lucky to join MCL.

3. What’s your future expectation for MCL?

To tell the truth, I am completely new to computer vision and machine learning. I feel really [...]

The segmentation research in our lab made a significant breakthrough these days. The paper “Robust Image Segmentation Using Contour-guided Color Palettes” by Xiang Fu, Chien-Yi Wang, Chen Chen, Changhu Wang and C.-C. Jay Kuo was accepted by ICCV 2015. In this paper, the contour-guided color palette (CCP) is proposed to efficiently integrate contour and color cues of an image. To find representative colors of an image, color samples along long contours between regions, similar in spirit to machine learning methodology that focus on samples near decision boundaries, are collected to achieve an image-dependent color palette. This color palette provides a preliminary segmentation in the spatial domain, which is further fine-tuned by post-processing techniques such as leakage avoidance, fake boundary removal, and small region mergence. While CCP offers an acceptable standalone segmentation result, it can be further integrated into the framework of layered spectral segmentation to produce a more robust segmentation.

For more details, please refer to our paper that will be published soon. The latest code of this paper can be downloaded here. It is written in MATLAB, and has been tested under 64-bit Windows, Linux, and Mac OSX.