Location:  Double Tree By Hilton San Jose, 2050 Gateway Place, San Jose, CA 95110

Time: Dec 13, 2016 Tuesday, 1:30pm - 5:30pm         Agenda        See who's attended                       

Jan P. Allebach

Professor, Electrical and Computer Engineering, Purdue University

Bio: Jan P. Allebach is Hewlett-Packard Distinguished Professor of Electrical and Computer Engineering at Purdue University. Allebach is a Fellow of the IEEE, the National Academy of Inventors, the Society for Imaging Science and Technology (IS&T), and SPIE. He was named Electronic Imaging Scientist of the Year by IS&T and SPIE, and was named Honorary Member of IS&T, the highest award that IS&T bestows. He has received the IEEE Daniel E. Noble Award, the IS&T/OSA Edwin Land Medal, and is a member of the National Academy of Engineering. He currently serves as an IEEE Signal Processing Society Distinguished Lecturer (2016-2017).

Title: Image Quality: Beauty is in the Eyes of the Beholder  Watch Video

Abstract: As imaging becomes increasing pervasive in our daily lives, understanding image quality is more important than ever. The final mediator of image quality is the human viewer. Yet, surprisingly, the image quality measure that is currently most widely used SSIM, is not explicitly based on any aspects of the human visual system, although it is possible to establish a monotonic relationship between SSIM scores and Mean Opinion Scores of human observers. In this talk, I will discuss two image quality models recently developed in our laboratory that are explicitly based on aspects of the human visual system. The first model is intended for still images, and among other aspects, accounts for the inhibition of masking by visually recognizable structure in the image. The second model is intended for video, and accounts for the color-dependence of artifact visibility in compressed video sequences, as mediated by spatial activity and saliency. For both models, we present psychophysical data to validate the advantages of our methods, as judged by human observers.

Dick C.A. Bulterman

Professor and Chair of the Department of Computer Science, Vrije Universiteit Amsterdam

Research Fellow, CWI Amsterdam

Bio: Dick Bulterman is Chair of the Department of Computer Science at the Vrije Universiteit Amsterdam and Research Fellow at CWI, the Dutch national institute for research in computer science and mathematics. He is also chair of ACM SIGWEB and an active member of ACM SIGMM. Bulterman is former CEO and President of FXPAL, the Fuji Xerox research center in Palo Alto, California. He is co-general chair of ACM TVX 2017 (in Hilversum) and co-general chair of IEEE ICSC 2017. He received his PhD from Brown University.

Title: Measuring and Manipulating Audiences: A Personal Reflection  Watch Video

Abstract: Understanding the emotional reactions of audiences to a wide range of content types is an important area of research. In this talk, I provide a personal reflection on various approaches to modeling, quantifying and understanding audience behavior based on a broad range of evaluation techniques. Using results from a study of the Heineken Weasel television commercial as a backdrop, I provide an overview of evaluation approaches and their impact in long-term and real-time evaluation. The main contribution is a personal reflection on audience evaluation based on multi-situation affinity with the area.

Mohan Kankanhalli

Professor of Computer Science, National University of Singapore

Bio: Mohan Kankanhalli is Provost's Chair Professor of Computer Science at the National University of Singapore (NUS). He is also the Dean of NUS School of Computing. Before becoming the Dean in 2016, he was the NUS Vice Provost (Graduate Education) during 2014-2016 and Associate Provost during 2011-2013. Mohan obtained his BTech from IIT Kharagpur and MS & PhD from the Rensselaer Polytechnic Institute.

His current research interests are in Multimedia Computing, Information Security, Image/Video Processing and Social Media Analysis. He is the Director of SeSaMe - the Centre for "Sensor-enhanced Social Media" (

Mohan is on the editorial boards of several journals including the ACM Transactions on Multimedia, Springer Multimedia Systems Journal, Pattern Recognition Journal and Springer Multimedia Tools & Applications Journal. He is a Fellow of IEEE.

Title: Fusing Physical and Social Sensors for Situation Awareness  Watch Video
Abstract: With the spread of physical sensors and social sensors, we are living in a world of big sensor data. Though they generate heterogeneous data, they often provide complementary information. Combining these two types of sensors together enables sensor-enhanced social media analysis which can lead to a better understanding of dynamically occurring situations. 
We utilize event related information detected from physical sensors to filter and then mine the geo-located social media data, to obtain high-level semantic information. Specifically, we apply a suite of visual concept detectors on video cameras to generate "camera tweets" and develop a novel multi-layer tweeting cameras framework. We fuse "camera tweets" and social media tweets via a unified matrix factorization model. We perform matrix factorization on a spatial-temporal situation matrix formed from physical sensors signals, incorporating the surrounding social content to exploit a set of latent topics that can provide explanations for the concept signal strengths. We have tested our method on large scale real data including PSI stations data, traffic CCTV camera images and tweets for situation prediction as well as for filtering noise from events of diverse situations. The experimental results show that the proposed approach is effective.

Ilija Hadzic

Distinguished Member of Technical Staff, Nokia Bell Labs

Bio: Ilija Hadzic is the Distinguished Member of Technical Staff in Nokia Bell Labs (formerly the Bell Labs of Alcatel-Lucent and Lucent Technologies). He received his Ph.D degree from the University of Pennsylvania in Electrical Engineering in 1999.

At Penn, Ilija's built one of the early systems that used runtime programmability of FPGA technology to dynamically modify the hardware layers of the network stack during the operation.

At Bell Labs, Ilija's work includes basic and applied research, as well as creating new technologies used in commercial products. The projects he worked on had applications in wireline and wireless access networks, metro-aggregation networks, synchronization systems, and (recently) cloud-based multimedia systems. Ilija's technical expertise includes hardware architectures, operating systems, embedded systems, GPU- and FPGA-accelerated computing architectures, and networked systems.

Ilija is has been an author of numerous technical papers and patents, as well as the lead implementor of many hardware and software systems that are now part of Nokia product portfolio. In recognition of his technical contributions, he was named the Bell Labs Fellow in 2010 and received the NJ R&D Council Edison Award in 2016.

Title: Network-attached I/O Devices  Watch Video

Abstract: In our daily lives we constantly interact with computing devices, we carry them around, we use them at work, and we have opinions about their hardware or operating systems. We are also forced to frequently upgrade the computing infrastructure that we own in face of increasing application demands. However, we should only care about the applications and I/O devices. In other words, as long as the application performs in a satisfactory manner and the I/O devices enable comfortable interaction with the application, the details of the platform are secondary.

In this talk, I will present an architecture and design of an experimental system in which I/O devices are the first-class network resources and the only physical devices accessible to the user. In this system, the primary information exchanged between the user's device and the computing infrastructure is the device-specific media and the network connection is turned into a "long cable". What, in a traditional computing system, is achieved by rewiring cables becomes manipulation of network flows.

I will describe interesting technical problems that we had to solve while implementing this system (mostly related to graphical interactive applications) and show a few short videos clips that demonstrate the capabilities of the system.

David Gonzalez-Arguirre

Research Scientist, Intel Labs

David Gonzalez is a scene understanding leading research scientist at Intel Labs in Oregon USA. His research domain is robot cognition and perception. He has a large publication and oral presentation track in top conferences and journals (ICRA, IROS, ICPR, Humanoids, Elsevier-journals, etc.) and best paper awards on robot vision and sensor fusion. David obtained his MSc. at the CINVESTAV in Guadalajara-Mexico. He also holds the title (Dr.-Ing.) doctor in engineering  with summa cum laude from the KIT (Karlsruhe Institute of technology, Karlsruhe -Germany) in the Humanoids and Intelligence Systems Lab. David has conducted research as Postdoctoral fellow in multiple EU projects at the Institute for Anthropometrics and Robotics at the KIT. He has 12 years of experience as scientist and R&D engineer.

Title: Towards ubiquitous ∞-Resolution 3D+ Modeling : Image Synthesis for Multimodal and Multi-cue Visual Analytics  Watch Video

A real-time, robust and flexible visual recognition system capable to cope with shape- and materials-diversity requires:
  • Hybrid representations
  • Multimodal recognition methods
  • Large and high quality training and validation  data sets

The solution to these problems have off-line and on-line phases. The focus of this presentation is placed on various elements of the off-line phase. The talk introduces a minimal-cost fully automatic method to generate 3D object models with unlimited resolution from a single aligned image for arbitrary revolution objects. The scope is focused on objects with complex (non-lambertian) materials that have intricate free-form revolution surfaces. The approach is solely based on a single 2D image extracting two cues: The object's external contour and internal skeleton. This method achieves unprecedented high quality 3D models with ultra-efficient serialization. Claims are supported by extensive evaluation using data sets with ground-truth metrics for volume, area, file-size, and execution time.


What's BAMMF?

BAMMF is a Bay Area Multimedia Forum series. Experts from both academia and industry are invited to exchange ideas and information through talks, tutorials, posters, panel discussions and networking sessions. Topics of the forum will include but not limited to emerging areas in vision, audio, touch, speech, text, sensors, human computer interaction, natural language processing, machine learning, media-related signal processing, communication, and cross-media analysis etc. Talks in the event may cover advancement in algorithms and development, demonstration of new inventions, product innovation, business opportunities, etc. If you are interested in giving a presentation at the forum, please contact us.

Our Sponsors:
PARC, a Xerox Company

Hewlett Packard