Three special sessions were selected to be held in conjunction with FG 2015. Information on the special sessions is provided below.
Analysis of mouth motion for speech recognition and speaker verification
Description: Mouth motion constitutes a large part of our non-rigid facial motion and contains important information that allows machines to recognize speech when audio is corrupted or even inaccessible, to carry out identity verification and liveness detection at the same time, and to analyze mouth non-manuals for automatic sign language recognition. Despite the apparent motivation, the problem of mouth motion analysis is relatively under-studied. This partly is due to the multidisciplinary nature of the problem, and partly because of our limited understanding of the visual aspects of speech and mouth non-manuals.
Research into mouth motion analysis has been gaining momentum in recent years thanks to advances in computer vision. We have seen effective and efficient solutions to fundamental problems such as face detection and facial landmark localization and various techniques developed for modeling high-dimensional data which could be adapted to tackle the problem. This special session aims to push forward the state of the art of the research. Possible topics of interest include but are not limited to:
- Visual-based speech recognition
- Audiovisual speech recognition
- Visual-based speaker verification
- Audio/visual fusion for speaker verification and liveness detection
- Mouth non-manuals for automatic sign language recognition
Face and gesture recognition in forensics
Organizers: Julian Fierrez, Peter K. Larsen, (co-orgnized by COST Action IC 1106)
Papers due: 1 December 2014
CMT paper submission: Link (Select Track: “Special session on Face and Gesture Recognition in Forensics”)
Description: As part of the 11th IEEE international Conference on Automatic Face and Gesture Recognition (FG2015) several evaluations (competitions/challenges) were organized. Specifically, the following evaluations were accepted for FG2015: THE VIDEO PERSON RECOGNITION EVALUATION (Organized by Ross Beverdige, Bruce Draper, Patrick J. Flynn and Jonathon Phillips), THE KINSHIP VERIFICATION IN THE WILD EVALUATION (Organized by Jiwen Lu), THE AGE PROGRESSION COMPETITION (Organized by Andreas Lanitis and Nicolas Tsapatsoulis), and THE FINE-GRAINED EVALUATION ON FACE DETECTION IN THE WILD (Organized by Junjie Yan, Bin Zhen, Zhen Lei and Stan Z. Li). The results of these evaluations will be presented in the Special session on Evaluations.
Session on Previously Published Journal Papers
Authors of published journal articles on the topics related to FG 2015 will have the opportunity to present their work in the conference, subject to space availability and approval by the FG Technical Program Chairs. Journal papers published on or after January 1, 2013 and journal manuscripts accepted on or before December 31, 2014 will be eligible for consideration.
- Copy of the journal paper
- One paragraph description of the technical significance of their work to the FG community
- Short biography of the presenting author
- (Optional) Link to an existing oral presentation by the presenting author (e.g., TechTV or YouTube)
- (Optional) Short video in which the presenting author describes the technical significance of the paper to the FG community
Authors, whose papers are selected, should register for the conference in order to present their work. The selected papers will not be reprinted nor archived by FG 2015.
Proposal submission: January 10, 2015
Notifications: February 6, 2015
Session Date: [to be decided]