|
POST-DOC SUBJECT : UNDERSTANDING AND
MODELING OF THE INFLUENCE OF AUDIO ON VISUAL PERCEPTION DURATION: 12-18 MONTHS FROM: ASAP DESCRIPTION: Technicolor (http://www.technicolor.com) provides technology, systems and
services to its Media & Entertainment clients involved in the different
components of the video chain (content creation, production, distribution and
access). Technicolor Research & Innovation (R&I) is the research arm of
the company. In particular, Technicolor R&I is constantly investigating
human perception in new application scenarios. The ambition is to understand
how users perceive multimedia content, derive innovative computational models
from this understanding, and propose new applications/services. Target
applications include cinema, television, gaming, communication... Technicolor R&I in Rennes, France,
offers a Post-Doc position in the area of audio-visual perception. Technicolor
has an existing visual attention model, which is a computational algorithm that
predicts the main regions-of-interest in images/videos by modeling the human
visual system. This model currently considers only the visual information.
However, video is rarely shown without an audio track and the presence of audio
may influence visual attention. The goal of this Post-Doc is to gain understanding
of the influence of audio on visual attention, identify relevant audio
features/information/cues, and develop an audio-visual computational attention
model. More specifically, the main goals/tasks of
this research work include: * A complete state-of-the-art in the field.
* The definition/set-up of audio equipment
to be used in a subjective test environment in order to conduct audio-video
experiments with human participants. * An implementation of a solution
(algorithm) within the existing model: development of the detection of audio
cues as well as their fusion with visual cues. * Conduct a user study (subjective
experiment) The successful candidate must have a PhD
(or soon), and strong knowledge in audio processing or audio perception.
Additional knowledge in image processing, human perception would be valuable.
The ideal candidate will also have experience in conducting subjective
experiments (setup/protocol). Since software development is expected in this
research work, good programming skills (C/C++) are required. The position is located in the Video
Processing & Perception Lab, in Rennes, France, and proposes an excellent
salary package. Applicants should submit a CV, recent list
of publications, a statement of research interests and examples of research
work achievements. Applications should be submitted
electronically to philippe.guillotel@xxxxxxxxxxxxxxx Dr. Quan HUYNH-THU Technicolor Research & Innovation email: quan.huynh-thu@xxxxxxxxxxxxxxx |