• Permalink Gallery

    Congratulations to three MCL PhD students for passing their defense

Congratulations to three MCL PhD students for passing their defense

Congratulations to Harshad Kadu, Joe Yuchieh Lin and Xiang Fu, who passed their defense last week! Following are their thesis abstracts and they also shared their PhD experiences.

Thesis: Advanced Techniques for Human Action Classification and Text Localization (Harshad Kadu)

The thesis contains two main research topics:

1) Automatic human action classification with mocap data

We propose a TSVQ based multi-resolution string representation scheme that transforms the time-series of human poses into codeword sequences. Temporal and spatial features extracted from these sequences are combined together using novel fusion methods to achieve superior performance.

2) Text localization in natural scene images

Stable extremal region operator detects regions of interest in the proposed multi-stage incremental region classification framework. Geometric filtering, context-based text grouping and ensemble classifier stages, remove false positives and group related text regions into words.


Thesis: Experimental Design and Evaluation Methodology for Human-Centric Visual Quality Assessment (Joe Yuchieh Lin)
The problem of human-centric visual quality assessment (VQA) is extensively studied in this thesis. Our study includes three major topics: 1) design of a dataset for streaming video quality assessment, 2) development of a new and effective video quality assessment index, 3) exploration of a new methodology for human visual quality assessment based on the notion of just-noticeable-differences (JND).

Thesis: Advanced Visual Segmentation Techniques: Algorithm Design and Performance Analysis (Xiang Fu)

Two research topics are covered in the dissertation:

1) How to interactively represent and segment the object(s) in a video?

An interactive video object segmentation framework is proposed to handle complex and diverse scenes with large motion and occlusions, which outperforms the state-of-the-art Adobe After Effect.

2) How to reliably generate automatic image segmentation?

Region-Dependent Spectral Graph is designed to fuse contour, surface, and depth cues according to the type of regions. Depth cue can further merge the regions in the textured area, which is never applied for segmentation before.

Contour-guided Color [...]

By |August 31st, 2015|News|Comments Off on Congratulations to three MCL PhD students for passing their defense|

MCL Member Attended CVPR 2015 in Boston

Chun-Ting Huang, a member of MCL, attended CVPR 2015 hosted in Boston from June 8th to 12th to present his work in Biometric Workshop. The conference was held in Hynes Convention Center, where researchers from all over the world gathered to present their state-of-the-art results. There were six oral sessions, five poster sessions, and two plenary talks given by Yann LeCun and Jack L. Gallant. Unlike last year, CVPR newly added CNN Architectures into this year’s oral session, showing that CNN has been widely recognized and accepted by computer vision society.

According to Chun-Ting, “there were thousands of attendees presented in this large-scale event, and many interesting ideas were brought up and discussed with different aspects. It is an amazing and fruitful experience to me. I believe every researcher who attended this event surely has learned more than he/she originally expected.”

Chun-Ting presented his work in Biometric Workshop, and his paper’s title was “TAEF: A Cross-Distance/Environment Face Recognition Method”. This system provides a cross-distance face alignment, a cross-environment face enhancement, and a two-stage coarse-to-fine approach to handle long distance faces. The performance of this method achieves 97% first rank accuracy on 150 meter visible-light images in the LDHF database.

By |June 14th, 2015|News|Comments Off on MCL Member Attended CVPR 2015 in Boston|
  • Permalink Gallery

    USC and PWICE Joint Project on Smart Glasses for Aircraft Maintenance

USC and PWICE Joint Project on Smart Glasses for Aircraft Maintenance

In Summer 2014, USC MCL and Pratt & Whitney Institute for Collaborative Engineering (PWICE) started a joint research project to build a remote collaboration system based on smart glasses. Mobile wearable devices have become popular nowadays. Smart glasses, among one of them, play an essential role in information capturing and visual enhancement. Particularly, in the sophisticated working environment for aircraft maintenance, workers have to retrieve and share information in real time with visual aids but may not have an easy access to a laptop or a tablet. In this project, we aim to develop a remote collaboration system based on smart glasses. With this remote collaboration system, each technician can use his/her smart glasses to communicate with experts/colleagues at a remote distance without interrupting the ongoing work.

In 2014, USC MCL has developed a prototype of remote collaboration system, live streaming app based on Google Glass. Video streaming system on wearable devices has demonstrated the potential influence when smart glasses were used in the plant. In 2015 and 2016, USC MCL and PWICE will continue to build the remote collaboration system, focusing on developing the Quality of Service (QoS) and Quality of Experience (QoE) optimization for aircraft maintenance.

By |June 7th, 2015|News|Comments Off on USC and PWICE Joint Project on Smart Glasses for Aircraft Maintenance|
  • Permalink Gallery

    MCLab Hosts a Visiting Student from Indian Institute of Technology Bombay

MCLab Hosts a Visiting Student from Indian Institute of Technology Bombay

The Viterbi-India program is jointly funded by the Indo US Science, Technology Forum (IUSSTF) and the USC VSoE. Each student in the Viterbi-India program will receive stipend and travel reimbursement. These students will come to USC and join one of the labs under EE department during May 20 and July 18, 2015.

After a competitive review process, there will be 20 spots for the Indian students in the program. Among those strong Indian students, Vikranth Reddy is the one who will join MCL in 2015 summer as an intern. We conducted a short interview with him about his previous experiences and future expectations.


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

I am Vikranth Dwaracherla, a third year undergraduate from IIT Bombay, India. I am currently pursuing my major in Electrical Engineering and minors in Computer Science. My previous research projects include neural networks, device simulations and image processing. I am very interested in the fields of computer vision and robotics. I have been working on a project on autonomous fruit collecting robot.

2.What was your first impression of USC and MCL?

USC is a wonderful place and provides opportunities for all people invariant of their interests. It is a wonderful place where people can pursue their goals. The members in the lab are all very kind and willing to help others. I would like to thank all the members of MCLab for the hospitality shown towards me.

3.What is your further expectation of being an MCL member?

I am very happy to be selected into the Viterbi- India Program and come to USC, one of the leading institutions in the field of Engineering and Technology. I am also very pleased to join MCLab, and to have the opportunity to work with [...]

By |May 31st, 2015|News|Comments Off on MCLab Hosts a Visiting Student from Indian Institute of Technology Bombay|

Congratulations to Xu (Becky) Qiu for Graduation with MSEE

Xu (Becky) Qiu, one of MCL members, graduated from USC on May 15, 2015. Congratulations to her and her families for her accomplishments in MCL and USC.

Xu (Becky) Qiu received her B.E. degree in Software Engineering from Shandong University (SDU), Jinan, China in 2011. Since 2013, she has been a research assistant in Media Communications Lab (MCL) at USC, advised by Prof. C.-C. Jay Kuo. Starting from Fall 2013, she served as a member of MCL Web Committee for a year, in charge of the alumni web page maintenances and updates. Becky was also the TA for a web design summer course for MCL interns in Summer 2014. In Fall 2014, she was the TA of Digital Image Processing (EE569). Since May 2014, she has been working on PWICE Google Glass Project as the team leader. On May 15, 2015, Becky Qiu got her M.S. degree in Electrical Engineering from University of Southern California (USC), Los Angeles.

By |May 24th, 2015|News|Comments Off on Congratulations to Xu (Becky) Qiu for Graduation with MSEE|

Five MCL Members Attended Viterbi PhD Hooding Ceremony

Five MCL members attended the Viterbi PhD hooding ceremony on Thursday, May 14, 2014, from 8:30-11:00 a.m. in the Bovard Auditorium. They were Sachin Chachada, Xiang Fu, Pang-Chang Lan, Sudeng Hu, and Joe Yuchieh Lin. Congratulations to them and their families for their accomplishments in completing their PhD program at USC.

Sachin Chachada received his B.E. degree in Electronics and Communication Engineering from S.R.K.N.E.C., RTM Nagpur University and an M.S. degree in Electrical Engineering from the University of Southern California (USC). Since 2009, he has been a member in the Media Communications Lab at USC, participating in the fields of statistical signal processing, and machine learning with applications to image and audio analysis. His dissertation, entitled “Environmental Sound Recognition: Classification and Retrieval,” discusses the algorithms that can be used to advance general audio understanding and management. His work demonstrates superior performance of ensemble learning algorithm for environmental audio classification, with use of classical, contemporary and a new set of time-frequency features. His work also includes novel algorithms for environmental sound retrieval for a large database, with promising applications for audio content management.

Xiang Fu received his B.S. degree in Electronic Engineering from Shanghai Jiao Tong University (SJTU), China in 2009, and his M.S. degrees in Electrical Engineering and in Computer Science, both from University of Southern California (USC), Los Angeles in 2011 and 2014, respectively. Since 2011, he has been pursuing his Ph.D. degree in Electrical Engineering with Media Communications Lab (MCL) at USC advised by Prof. C.-C. Jay Kuo, and has been working on various research areas including image/video segmentation, spectral clustering, visual tracking, object recognition, video surveillance, and machine learning. His dissertation, entitled “An Information Fusion Approach to Visual Data Segmentation”, discusses feature fusions [...]

By |May 17th, 2015|News|Comments Off on Five MCL Members Attended Viterbi PhD Hooding Ceremony|
  • Permalink Gallery

    USC and Netflix Joint Research to Be Presented at ICIP 2015 and ICME 2015

USC and Netflix Joint Research to Be Presented at ICIP 2015 and ICME 2015

Netflix ingest and encoding pipeline is a cloud-based platform that generates video encodes for the Netflix streaming service. Due to the large throughput of the system, automated video quality assessment of the source videos and the generated encodes is essential in ensuring the quality of experience of Netflix subscribers. Owing to the diversity of video content, traditional video quality assessment methods are not able to meet the need of Netflix video pipeline. A joint research team between USC MCL and Netflix is assembled to tackle this difficult problem.

A scalable solution of video quality assessment method is proposed by Joe Yuchieh Lin (USC), Eddy Chihao Wu (USC), Dr. Ioannis Katsavounidis (Netflix), Dr. Zhi Li (Netflix), Dr. Anne Aaron (Netflix), and Prof. C.-C. Jay Kuo (USC). This method is called Ensemble-Learning-based Video Quality Assessment Index (EVQA). EVQA adopts a frame-based learning mechanism to address the limited training data problem and fuses multiple image quality assessment indices to generate the final video quality score. The superior performance of the proposed EVQA index is demonstrated by experimental results conducted on both LIVE and MCL-V video databases.

The discussion of Netflix’s pipeline system, current solutions and remaining challenges will appear in ICIP 2015. The work of EVQA is accepted by ICME 2015 workshop on Cloud-based Media.

By |May 10th, 2015|News|Comments Off on USC and Netflix Joint Research to Be Presented at ICIP 2015 and ICME 2015|
  • Permalink Gallery

    MCL Research on Cross-Distance/Environment Face Recognition Accepted by CVPR Workshop on Biometrics

MCL Research on Cross-Distance/Environment Face Recognition Accepted by CVPR Workshop on Biometrics

Face recognition has been studied and developed for over decades. However, the long distance face recognition accompanied with automatic alignment is still a challenging topic due to its distorted and quality-degraded environment. This problem is also known as Face Recognition at A Distance (FRAD), which is a common issue in video surveillance applications.

A solution to FRAD has been proposed by MCL PhD student, Chun-Ting Huang, visiting scholar from UESTC, Professor Zhengning Wang, and Professor C.-C. Jay Kuo.  The method called Two-Stage Alignment/Enhancement Filtering (TAEF) system consists of three main components: a cross-distance face alignment technique, a cross-environment face enhancement technique, and a two-stage filtering system. In the first stage, the given probe image is adjusted and examined in coarse-scale for eliminating unlikely candidates, and then the procedure is conducted for every individual probe/gallery image pair for higher accuracy at the second stage. The first rank recognition rates of the TAEF method are 100%, 100% and 97% for 60-, 100- and 150-meter visible-light images in the LDHF database, respectively.

This work has been accepted for publication in the Computer Vision and Pattern Recognition (CVPR) 2015 Workshop on Biometrics held in Boston, June 11th, 2015.

By |May 3rd, 2015|News|Comments Off on MCL Research on Cross-Distance/Environment Face Recognition Accepted by CVPR Workshop on Biometrics|

New Year Greetings

2014 is a fruitful year for MCLab and we are looking forward to a wonderful 2015.

Wish all members happy new year!


Image credit: Latin Times

By |December 28th, 2014|News|Comments Off on New Year Greetings|

Professor Kuo Completed His Service as APSIPA President

Professor C.-C. Jay Kuo attended 2015 APSIPA Annual Summit Conference (ASC) in Siem Reap, Cambodia, December 9-12. APSIPA represents “Asia Pacific Signal and Information Processing Association”, which was founded in 2009. Professor Sadaoki Furui was the first President (2009-2012) while Prof. Jay Kuo was the second (2012-2014). During his tenure, Professor Kuo launched a major membership campaign through recruiting APSIPA Friend Labs. Right now, APSIPA has around 3200 e-members in the LinkedIn Social Network and 260 APSIPA Friend Labs.

Professor Kosin Chamnongthai of King Mongkut’s University of Technology (Thailand), Professor Hitoshi Kiya of Tokyo Metropolitan University (Japan) and Professor Kuo were General Co-Chairs of APSIPA ASC 2014. The conference location, Siem Reap, is the home to the incredible Angkor ruins, a sprawling World Heritage-listed complex of ancient temples, including Angkor Wat, Angkor Thom, etc. Professor Kuo met a couple of MCL alumni from Korea and Taiwan.

Professor Kuo completed his two-year term of APSIPA President immediately after this conference, and Dr. Haizhou Li of Singapore has succeed Professor Kuo to become the new President of APSIPA. Dr. Li is a world renowned expert and technical leader in speech processing and recognition.

By |December 21st, 2014|News|Comments Off on Professor Kuo Completed His Service as APSIPA President|