|Aleix Martinez, Ohio State University, USA. Deciphering the Face. Much progress has been made to understand human cognition. Yet, little is know about face perception. Facial expressions of emotions is a clear example of this limitation. It has been established how the underlying muscles move in response to felt emotions, but little is known on how these constructs are interpreted by the visual system and how to build robust computer vision systems that emulate human perception. In this talk, we will review the recent literature on this subject and proposes a model for the perception of facial expressions of emotion. [more...]|
|Greg Mori, Simon Fraser University, Canada. Learning structured models for recognizing human actions. The development of automatic methods for recognizing human actions is a challenging computer vision problem. Robust solutions to this problem would facilitate a variety of applications from image retrieval to improving safety in assisted living facilities. In this talk I will present work towards solving this problem via the learning of structured models. I will describe a model that uses a hidden Conditional Random Field (hCRF) to learn a representation for motion parts in conjunction with whole-body templates. Second, a variant of this model is used for treating body pose as a latent variable for action recognition. [more...]|
|Richard Bowden, Univ.
Surrey, UK. From activity to
language: learning to recognise the meaning of motion. Whether the task
is recognising an atomic action of an individual or their implied
activity, the continuous multichannel nature of sign language
recognition or the appearance of words on the lips, all approaches can
be categorised at the most basic level as the learning and recognition
of spatio-temporal patterns. The complexity of the problem to some
extend dictates the approaches used e.g. for simple gestures and
actions, holistic descriptors can work well but are less suitable for
the subtitles and complexities of sign. However, [more...]
|Graham Taylor, NYU, New-York. Tutorial on applications of Deep Learning methods to activity recognition. Recognition of human activity from video data is a challenging problem that has received an increasing amount of attention from the computer vision community in recent years. Currently the best performing methods at this task are based on engineered descriptors with explicit local geometric cues and other heuristics. In this tutorial, we will review a number of recently proposed methods that attempt to learn low and mid-level features for use in activity recognition. This includes deep and unsupervised feature learning methods such as convolutional networks, convolutional deep belief networks and other approaches [more...]|
|David Forsyth, University of Illinois at Urbana-Champaign. Understanding human activity. David Forsyth is well known for his work on human motion computing. He and co-authors have demonstrated the first robust, accurate human tracker that can reliably report the configuration of arms and legs and does not need to be started by hand. They showed that motion capture data can be rearranged to produce highly realistic animations of novel human motions, and demonstrated that close control of the nature of the motion was possible using annotations. They then demonstrated that this animation system can be linked to the output of this tracker, to obtain annotations describing human activities automatically from video.|
|Sudeep Sarkar, University of South Florida. Segmentation-robust representations, matching, modeling for sign language recognition. Distinguishing true signs from transitional, extraneous movements made by the signer as s/he moves from one sign to the next is a serious hurdle in the design of continuous Sign Language recognition systems. This problem is further compounded by the ambiguity of segmentation and occlusions, resulting in propagation of errors to higher levels. This talk will describe our experience with representations and matching methods, particularly those that can handle errors in low-level segmentation methods and uncertainties in segmentation of signs in sentences. [More...]|
Rutgers University, New Jersey.
language and human activity recognition. Dimitris Metaxas is a
Professor II (Distinguished) in the Division of Computer and
Information Sciences at Rutgers University where he conducts research
on the development of formal methods upon which computer vision,
computer graphics and medical imaging can advance synergistically. In
particular he is focusing on human body and shape motion analysis,
human surveillance, security applications, American Sign Language
recognition, behavior modeling and analysis and scalable solutions to
large and distributed sensor-based networks.
|Maja Pantic, Imperial College, London. Facial Behaviour Understanding. The facial behaviour is our preeminent means to communicating affective and social signals. This talk discusses a number of components of human facial behavior, how they can be automatically sensed and analysed by computer, what is the past research in the field conducted by the iBUG group at Imperial College London, and how far we are from enabling computers to understand human facial behavior. Maja Pantic is a Professor of Affective and Behavioral Computing and leader of the Imperial College of London, working on machine analysis of human non-verbal behavior and its applications to Human Computer Interaction. [More...]|
|Christian Vogler, Gallaudet University, Washington DC. Advances in Phonetics-based Sub-Unit Modeling for Transcription Alignment and Sign Language Recognition. Christian Vogler is the director of the Technology Access Program. He is a principal investigator within the Rehabilitation Engineering Research Center (RERC) on Telecommunications Access, with a particular focus on the accessibility of web conferencing and telecollaboration systems. In his role at the RERC, he is involved in bringing consumers and industry together on accessibility issues, as well as developing prototype technologies for improving the accessibility of such systems. [More...]|
Carnegie Mellon University.
motion detection and understanding using both 2D and 3D setups. Takeo Kanade is the
U. A. and Helen Whitaker University Professor of Computer Science and
Robotics and the director of Quality of Life Technology Engineering
Research Center at Carnegie Mellon University. He works in multiple
areas of robotics: computer vision, multi-media, manipulators,
autonomous mobile robots, medical robotics and sensors. He has written
more than 300 technical papers and reports in these areas, and holds
more than 20 patents. He has been the principal investigator of more
than a dozen major vision and robotics projects at Carnegie Mellon.
Monday, June 20, 2011
Colorado Ballroom B
papers are published in IEEE Explore. For your convenience the workshop
participants can access preprints from this page. Login: gesture.
9:00 am. Christian Vogler, Gallaudet University, Washington DC. Advances in Phonetics-based Sub-Unit Modeling for Transcription, Alignment and Sign Language Recognition. Pascal2 best paper award. Co-authors: Vassilis Pitsikalis, Stavros Theodorakis, and Petros Maragos [Abstract][Preprint][Slides pptx][Sound] .
9:45 am. Sudeep Sarkar. University of South Florida. Segmentation-robust Representations, Matching, and Modeling for Sign Language Recognition. Co-authors: Barbara Loeding, Ruiduo Yang, Sunita Nayak, Ayush Parashar. [Abstract][Preprint][Slides].
10:15 am. Break [coffee, snacks provided]
10:30 am. Dimitri Metaxas, Rutgers University, New Jersey. Sign language and human activity recognition. [Slides].
11:00 am. Richard Bowden, Univ. Surrey, UK. From activity to language: learning to recognise the meaning of motion. [Abstract][Slides].
11:30 am. Aleix Martinez, Ohio State University, USA. Deciphering the Face. [Abstract][Preprint][Slides].
12:00-2:00 pm: Poster session and lunch (included) served in the foyer/hallway and Summit Ballroom (4th floor). Finger food will be provided so the participants can talk and view the posters during lunch time.
Afternoon: Gestures and actions
2:00 pm. David Forsyth, University of Illinois at Urbana-Champaign. Understanding human activity. [Slides].
2:45 pm. Graham Taylor, NYU, New-York. A tutorial on deep and unsupervised feature learning for activity recognition. [Abstract][Slides].
3:15 pm. Maja Pantic, Imperial College, London. Designing Frameworks for Automatic Affect Prediction and Classification in Dimensional Space. Co-authors: Mihalis A. Nicolaou and Hatice Gunes. [Abstract][Preprint][Slides].
3:45 pm. Break.
4:00 pm. Greg Mori, Simon Fraser University, Canada. Learning structured models for recognizing human actions. [Slides]
4:30 pm. Takeo Kanade, Carnegie Mellon University. Body motion detection and understanding using both 2D and 3D setups.
5:00 pm. Discussion.
5:30 pm. Adjourn.
7:00 pm. Dinner invitation for the invited speakers and the organizers at Fratelli Ristorante. The other participants may join at their own expenses (please contact the organizers).
Each presenter will
have one side of a poster board. Poster size = 4'x 8' max (4'
tall x 8' wide or 4' veritcal x 8' horizontal) see picture.
6DMG: A New 6D Motion Gesture Database. Mingyu Chen, Ghassan AlRegib, and Biing-Hwang Juang, Georgia Institute of Technology, Atlanta, Georgia, U.S.A. [Extended summary]
Invariant Facial Emotion Classification using 3D Constrained Local
Model and 2D Shape Information. Laszlo A. Jeni, University of
Tokyo, Japan, Hideki Hashimoto, Chuo University, Japan, András
Eotvos Lorand University, Hungary. [Extended summary]
An Optical Flow-based Action Recognition Algorithm. Upal Mahbub, Hafiz Imtiaz, and Md. Atiqur Rahman Ahad. Bangladesh University of Engineering and Technology. [Extended summary]
SURF- and Optical
Flow-based Action Recognition with Outlier Management. Atiqur
Rahman Ahad, J. Tan, H. Kim, S.
Kyushu Institute of Technology, Japan. [Extended
Collection (GDC) software for the Gesture Recognition Challenge.
Isabelle Guyon, Clopinet, California and Vassilis Athitsos, University
of Texas at Arlington. [Documentation][Download GDC[
Unsupervised and Transfer Learning
Challenge. Just ended.
active learning and experimental design workshop. With guest
speakers Donald Rubin, Burr Settles, and David Jensen.
special session on autonomous and incremental learning and
competition on active learning.
2009 causality and time series mini-symposium. Featuring a
memorial lecture of Clive Granger by Halbert White.
NIPS 2008 causality workshop: objectives and assessment. The second challenge in causality organized by the causality workbench.
WCCI 2008 causation and prediction challenge. A first activity of the causality workbench.
NIPS 2006 workshop on causality and feature selection. A first workshop on causality..
IJCNN 2007 Agnostic learning vs. Prior knowledge challenge. “When everything fails, ask for additional domain knowledge” is the current motto of machine learning. Therefore, assessing the real added value of prior/domain knowledge is a both deep and practical question.The participants competed in two track: the “prior knowledge track” for which they had access to the raw data and information about the data, and the “agnostic learning track” for which they had access to preprocessed data with no knowledge of the identity of the features.
WCCI 2006 performance prediction challenge. “How good are you at predicting how good you are? 145 participants tried to answer that question. Cross-validation came very strong. Can you do better? Measure yourself against the winners by participating to the model selection game.
NIPS 2003 workshop on feature extraction and feature selection challenge. We organized a competition on five data sets in which hundreds of entries were made. The web site of the challenge is still available for post challenge submissions. Measure yourself against the winners! See the book we published with a CD containing the datasets, tutorials, papers on s.o.a. methods.
Pascal challenges: The Pascal network is sponsoring several challenges in Machine learning.
Data mining competitions:
A list of data mining competitions maintained by KDnuggets, including the well known KDD cup.
of data sets for machine learning:
A rather comprehensive list maintained by MLnet.
UCI machine learning repository: A great collection of datasets for machine learning research.
DELVE: A platform
Critical Assessment of Microarray Data Analysis, an annual conference on gene expression microarray data analysis. This conference includes a context with emphasis on gene selection, a special case of feature selection.
International Conference on Document Analysis and Recognition, a bi-annual conference proposing a contest in printed text recognition. Feature extraction/selection is a key component to win such a contest.
Text Retrieval conference, organized every year by NIST. The conference is organized around the result of a competition. Past winners have had to address feature extraction/selection effectively.
In conjunction with the International Conference on Pattern Recognition, ICPR 2004, a face recognition contest is being organized.
An important competition in protein structure prediction called Critical Assessment of
Techniques for Protein Structure Prediction.
We are very grateful to all the reviewers:
Aleix Martinez, Ohio State University, USA
David W. Aha, Naval Research Laboratory, USA
Abe Schneider, Knexus Research, USA
Jeffrey Cohn, Carnegie Mellon University, USA
Martial Hebert, Carnegie Mellon University, USA
Dimitris Metaxas, Rutgers, New Jersey, USA
Christian Vogler, ILSP Athens, Greece
Sudeep Sarkar, University of South Florida, USA
Graham Taylor, NYU, New-York, USA
Andrew Ng, Stanford Univ., Palo Alto, CA, USA
Andrew Saxe, Stanford Univ., Palo Alto, CA, USA
Quoc Le, Stanford Univ., Palo Alto, CA, USA
David Forsyth, University of Illinois at Urbana-Champaign, USA
Maja Pantic, Imperial College, London
Philippe Dreuw, RWTH Aachen University, Germany
Richard Bowden, Univ. Surrey, UK
Fernando de la Torre, Carnegie Mellon University, USA
Paulo Gotardo, Ohio State University, Ohio, USA
Carol Neidle, Boston University, MA, USA
Trevor Darrell, UC Berkeley/ICSI, Berkeley, California, USA
Greg Mori, Simon Fraser University, Canada
Matthew Turk, UC Santa Barbara, USA
Atiqur Rahman Ahad, Faculty of Engineering, Kyushu Institute of Technology, Japan
Mingyu Chen, Georgia Institute of Technology, USA
Wenhui Xu, Georgia Institute of Technology, USA
Jesus-Omar Ocegueda-Gonzalez, University of Houston, USA
Thomas Kinsman, Rochester Institute of Technology, USA
András Lőrincz, Eötvös Loránd University, Budapest, Hungary
Upal Mahbub, Bangladesh University of Engineering, Bangladesh
Subhransu Maj, UC Berkeley, CA, USA
Lubomir Bourdev, UC Berkeley, CA, USA
Vassilis Pitsikalis, NTUA, Greece
gesture@ clopinet . com.
We are grateful to the DARPA Deep
Learning program, the Naval Research Laboratory, and the Pascal2
European network for their support.