The UBMM Lab at State University of New York at Buffalo

Wenyuan Yin

wyin4@buffalo.edu
Curriculum Vitae

Research Interests

Photo quality assessment with context and crowdsourced photos: learning scene dependent aesthetic model from related images in social media community for photo quality assessment scheme by utilizing context and image content

Socialized mobile photography: Developing crowdsourced approach to learn relevant photography rules from social media to guide mobile users capture high quality photos via mobile devices

User centric media adaptation for mobile device: developing user centric media adaptation schemes by analyzing media semantics and user interests to improve mobile user viewing experience

Social media snippet generation for mobile browsing: developing automatic social media snippet generation scheme to present key information of social posts based on visual perception principles to maximize information perception on mobile device

Honors and Awards

1.    VCIP 2012 Best Student Paper Award, 2012
2.    Microsoft Research Asia Excellent Intern Award-'Star of Tomorrow', 2012
3.    Outstanding Graduate Student, 2006
4.    Outstanding Student Award (2 winners/340 students) for 3 consecutive years, 2003-2005

Educational Background

State University of New York at Buffalo, United States Aug., 2008 – Present
Ph.D Candidate, Computer Science and Engineering Department
Thesis Topic: Mobile Multimedia: from Acquisition to Adaptation with Semantics, Context and Social Information
Advisor: Prof. Chang Wen CHEN
GPA: 3.9/4.0

Nanjing University of Science and Technology, China Sep.,2006 – Jul., 2008
M.S., School of Computer Science and Technology
GPA: 90.0/100.0

Nanjing University of Science and Technology, China Sep., 2002 – Jul., 2006
B.E., School of Computer Science and Technology
GPA: 88.4/100

Projects

Photo quality assessment with context and crowdsourced photos: Most existing approaches to photo quality assessment have predominantly focused on image content itself, while ignoring various contexts such as the associated geo-location and timestamp. However, such a universal aesthetic assessment model may not work well with significantly different contexts, since the photography rules are always scene and context dependent. We leverage the context information associated with photos for visual quality assessment utilizing relevant photos in social media community. In particular, the research mainly focuses on the following aspects:

Leaning Scene-Dependent Aesthetic Models (SDAM) to assess photo quality from contextual searched related photos, by jointly leveraging the context and visual content.
Transferring related aesthetic rules from auxiliary photos of the same scene category from other locations to overcome the insufficient data issue for SDAM learning.

Socialized mobile photography : The popularity of mobile devices equipped with various cameras has revolutionized modern photography. However, taking a high quality photograph via mobile device remains a challenge for mobile users. We investigate a photography model to assist mobile users in capturing high quality photos by using both the rich context available from mobile devices and crowdsourced social media. The research mainly focuses on the following aspects:

Addressing the complex scene and context dependent photography challenge.
Suggesting the optimal view enclosure (composition) given a wide view of scene by view cluster discovering and view specific composition learning .
Recommending appropriate camera parameters (aperture, ISO, and exposure time) considering the complicated effects of shooting content and various context by metric learning.

User Guided Semantic Media Adaptation for Mobile Device : With user’s guidance, developing semantic based media adaptation schemes to accommodate the variation of resource constraints and limitation of capabilities of mobile devices in heterogeneous networks. We explore visual cues and semantic concepts of consumer media and design the semantic based adaptation scheme to achieve optimal adaptation result to improve end users’ perceptual experience under various resource constraints. In particular, the research mainly focuses on the following aspects:

Investigating and developing semantic concept extraction methods in media contents for subsequent semantic adaptation.
User guided seamless integration of mobile user supplied semantic information with low level image features to generate perceptually optimized and semantically important regions for adaptation.
Intelligent content selection and adaptation with the user display parameters to provide mobile user optimal perception experience.

Social media snippet generation for mobile browsing : More and more people are using their mobile devices to enjoy social media contents on the move. However, mobile display constraints bring challenges for presenting the rich media content on the limited screens. We propose an innovative system to automatically generate social media snippet for efficient mobile browsing. In particular, the research mainly focuses on the following aspects:

Excerpting the most salient and dominant elements (both textual and visual contents) from the original media content.
Optimally selecting and composing the elements into a snippet based on human visual perception principles, aesthetic rules considering mobile display constraint to actively push important information to mobile users for efficient interested information discovering.

Publications

Journal Papers

W. Yin, T. Mei, C. W. Chen, “Assessing Photo Quality with Context and Crowd-sourced Photos”, to be submitted to IEEE Transactions on Multimedia

W. Yin, T. Mei, C. W. Chen and S. Li, " Socialized Mobile Photography: Learning to Photograph with Social Context via Mobile Devices ", IEEE Transactions on Multimedia, 16(1): 184-200, 2014 (has been selected as featured article in Special Technical Community on Social Networks April 2014)

W. Yin, J. Luo, C. W. Chen, “Event-based Semantic Image Adaptation for User-centric Mobile Display Devices,” IEEE Transactions on Multimedia, 13(3): 432-442, 2011

Conference Papers

W. Yin, T. Mei, C. W. Chen, " Automatic Generation of Social Media Snippets for Mobile Browsing ," ACM International Conference on Multimedia, Oct, 2013 (full paper)

W. Yin, T. Mei, C. W. Chen, “ Assessing Photo Quality With Geo-Context And Crowdsourced Photos,” IEEE International Conference on Visual Communications and Image Processing, Nov, 2012 (Best Student Paper Award)

W. Yin, T. Mei, C. W. Chen, “Crowdsourced Learning To Photograph Via Mobile Devices,” IEEE International Conference on Multimedia & Expo, July, 2012

W. Yin, X. Zhu, C. W. Chen, “Contemporary Ubiquitous Media Services: Content Recommendation and Adaptation,” Workshop on Pervasive Communities and Service Clouds, IEEE International Conference on Pervasive Computing and Communications, March, 2011

W. Yin, J. Luo, C. W. Chen, “Semantic Image Adaptation for User-centric Mobile Display Devices,” IEEE Communications Society Multimedia Communications Technical Committee E-Letter, invited, Vol.6, No.1, January, 2011

W. Yin, J. Luo, C. W. Chen, “User Guided Semantic Image Adaptation for Mobile Display Devices,” IEEE International Conference on Multimedia & Expo, July, 2010

W. Yin, J. Luo, C. W. Chen, “Semantic adaptation of consumer photo for mobile device access,” IEEE International Symposium on Circuits and Systems 2010, May, 2010