|
Theme: HCI and applications - Chairperson: Barbara Peskin, ICSI, Berkeley
|
|
| 08:30 | Invited talk: Bill Buxton - Buxton Design, Mountains, Exploration, Education, Rich Media and Design |
|
|
|
| 09:15 | Simon Tucker, Steve Whittaker - Univ. of Sheffield, UK,
Accessing Multimodal Meeting Data: Systems, Problems and Possibilities
|
|
|
|
|
(noisy)
|
slides
|
all files
|
"Summarize this hour long meeting in five minutes [06:43s]"
|
|
| 09:35 | Pierre Wellner, Mike Flynn, Mael Guillemot - IDIAP, CH Browsing Recorded Meetings With Ferret |
| |
|
| 09:55 | Dennis Reidsma, Rutger Rienks, Natasa Jovanovic- Univ. of Twente, NL Meeting Modelling |
| |
|
|
|
|
|
Theme: Structuring and interaction - Chairperson: Pierre Wellner, IDIAP, CH
*** Hint for this session: play the audio-visual presentation while browsing the slides. ***
|
|
| 11:00 | Invited talk: Rafah Hosn - IBM Research, A Programming Model for Next Generation Multimodal Applications |
|
|
| 11:45 | Denis Lalanne, Rolf Ingold, Didier von Rotz, Ardhendu Behera, Dalila Mekhaldi, Andrei Popescu-Belis - Univ. of Fribourg, CH , Using Static Documents as Structured and Thematic Interfaces to Multimedia Meeting Archives |
|
|
| 12:05 | Agnes Just, Sebastien Marcel, Olivier Bernier - IDIAP, CH , Recognition of Isolated Complex Mono- and Bi-Manual 3D Hand Gestures using Discrete IOHMM |
|
|
| 12:25 | Nicolas Moënne-Loccoz, Bruno Janvier, Stéphane Marchand-Maillet, Eric Bruno - Univ. of Geneva, CH, An Integrated Framework for the Management of Video Collection |
|
|
|
|
|
|
Theme: Speech Processing - Chairperson: Steve Renals, Univ. of Edinburgh, UK
|
|
| 14:00 | Invited talk: Stephen Cox - University of East Anglia, Confidence Measures in Speech Recognition |
|
|
| 14:45 | Chuck Wooters, Nikki Mirghafori, Andreas Stolcke, Tuomo Pirinen, Ivan Bulyko, Dave Gelbart, Martin Graciarena, Scott Otterson, Barbara Peskin, Mari Ostendorf - ICSI, Berkeley, The 2004 ICSI-SRI-UW Meeting Recognition System |
|
|
| 15:05 | Mathew Magimai Doss, Hervé Bourlard - IDIAP, CH, On the Adequacy of Baseform Pronunciations and Pronunciation Variants |
|
|
| 15:25 | Qifeng Zhu, Barry Chen, Nelson Morgan, Andreas Stolcke - ICSI, Berkeley, Tandem Connectionist Feature Extraction for Conversational Speech Recognition |
|
|
|
|
|
 up | Tuesday, June 22, 2004 |
|
Theme: Interaction and teleconferencing - Chairperson: Jean-Philippe Thiran, EPFL, CH |
|
| 08:30 | Invited talk: Jonathan Foote - FX Palo Alto Laboratory, Immersive Conferencing Directions at FX Palo Alto Laboratory |
|
|
|
|
|
slides
|
all files
|
"If you could save people from having to watch a boring meeting [08:40s] ..."
|
|
| 09:15 | Invited talk: Mats Ljungqvist - EU, EU research initiatives in multimodal interaction |
|
|
| 09:35 | Iain McCowan, Daniel Gatica-Perez, Samy Bengio, Darren Moore, Hervé Bourlard - IDIAP, CH, Towards Computer Understanding of Human Interactions |
|
|
|
|
|
slides
|
all files
|
"How can we model the underliying process that is generating the observations [2:14]"
|
|
| 09:55 | Max Froumentin - W3C, FR, Zakim - A multimodal sofware system for large-scale teleconferencing |
|
|
|
|
|
|
Theme: Multimodal processing - Chairperson: Daniel Gatica-Perez, IDIAP, CH
|
|
| 11:00 | Invited talk: Nuria Oliver - Microsoft Research, S-SEER: A Multimodal Office Activity Recognition System with Selective Perception |
|
|
| 11:45 | Tue Lehn-Schiøler, Lars Kai Hansen, Jan Larsen - Technical Univ. of Denmark, DK, Mapping from Speech to Images Using Continuous State Space Models |
|
|
| 12:05 | Ofer Dekel, Joseph Keshet, Yoram Singer - Hebrew University, Israel, An Efficient Online Algorithm for Hierarchical Phoneme Classification |
|
|
| 12:25 | Julien Meynet, Vlad Popovici, Jean-Philippe Thiran, EPFL, CH, Mixture of SVMs for Face Class Modeling |
|
|
|
|
|
|
Theme: Dialogue Management - Chairperson: Nelson Morgan, ICSI, Berkeley
|
|
| 14:00 | Invited talk: Jordan Cohen - VoiceSignal, Experiences with mobile device Multimodal applications |
|
|
| 14:45 | Oliver Lemon, James Henderson - Univ. of Edinburgh, Scotland, Machine Learning and the information state update approach to dialogue management |
|
|
| 15:05 | Andrei Popescu-Belis, Alexander Clark, Maria Georgescul, Sandrine Zufferey, Denis Lalanne - Univ. of Geneva, CH, Shallow dialogue processing using machine learning algorithms (or not) |
|
|
|
|
|
slides
|
all files
|
"The algorithm implemented is based on anaphora that comes from linguistics [15:00]"
|
|
| 15:25 | Agnes Lisowska, Martin Rajman, Trung H. Bui, Univ. of Geneva, EPFL, CH, ARCHIVUS: A System for Accessing the Content of Recorded Multimodal Meetings |
|
|
|
|
|
| up | Wednesday, June 23, 2004 |
|
Theme: Emotion & tracking - Chairperson: Jean Carletta, Univ. of Edinburgh, UK
|
|
| 08:30 | Invited talk: Roddy Cowie - Queen's University, Belfast, Piecing together the emotion jigsaw |
|
|
|
|
|
slides
|
all files
|
"some things are drawing our feelings, our attention [9:24]"
|
|
| 09:15 | Themis Balomenos, Amaryllis Raouzaiou, Spiros Ioannou, Athanasios Drosopoulos, Kostas Karpouzis, Stefanos Kollias - Technical Univ. of Athena, Greece, Emotion Analysis in Man-Machine Interaction Systems |
|
|
| 09:35 | Philipp Zehnder, Esther Koller-Meier, Luc Van Gool - ETHZ, CH, A Hierarchical System for Recognition, Tracking and Pose Estimation |
|
|
| 09:55 | Santiago Venegas, Gianluca Antonini, Jean Philippe Thiran, Michel Bierlaire - EPFL, CH, Automatic pedestrian tracking using discrete choice models and image correlation techniques |
|
|
|
|
|
|
Theme: Speech processing - Chairperson: Phil Green, Univ. of Sheffield, UK
|
|
| 11:00 | Invited talk: Yorick Willks - Sheffield University, Artificial Companions |
|
|
|
|
|
slides
|
all files
|
"Noone can avoid machine learning, there is no doubt about that [23:35]"
|
|
| 11:45 | Jean Carletta, Jonathan Kilgour - Univ. of Edinburgh, The NITE XML Toolkit meets the ICSI Meeting Corpus: import, annotation, and browsing |
|
|
| 12:05 | Barry Chen, Qifeng Zhu, Nelson Morgan - ICSI, Berkeley, Long-Term Temporal Features for Conversational Speech Recognition |
|
|
|
|
|
slides
|
all files
|
"We found from psycho-accoustics that there is information spread over time [3:00]"
|
|
| 12:25 | Harald Romsdorfer, Beat Pfister, René Beutler - ETHZ, CH, A Mixed-lingual Phonological Component in Polyglot TTS Synthesis |
|
|
|
| |
Support: 
Questions, comments, bug reports, and feedback very welcome.
|
Theme categories
