Presenter: Nam Le
EUMSSI Team at the MediaEval Person Discovery Challenge 2016 In Working Notes Proceedings of the MediaEval 2016 Workshop, Hilversum, Netherlands, October 20-21, CEUR-WS.org (2016) by Nam D. Le, Sylvain Meignier, Jean-Marc Odobez
Paper: http://ceur-ws.org/Vol-1739/MediaEval_2016_paper_46.pdf
Video: https://youtu.be/axt_PxhIrJ4
Abstract: We present the results of the EUMSSI team’s participation in the Multimodal Person Discovery task. The goal is to identify all people who simultaneously appear and speak in a video corpus. In the proposed system, besides improving each modality, we emphasize on the ranking of multiple results from both audio stream and visual stream.
MediaEval 2016 - EUMSSI Team at the MediaEval Person Discovery Challenge
1. EUMSSI
team
at
the
MediaEval
Person
Discovery
Challenge
2016
Nam
Le,
Jean-‐Marc
Odobez,
Sylvain
Meignier
{nle,
odobez}@idiap.ch
sylvain.meignier@univ-‐lemans.fr
3. Video
OCR
and
NER3
07/12/2016
Original
Image
Text region
detection
Text
extraction
Text
recognition
Hypothesis
merging
• Multiple
image
segmentations
of
the
same
region
è all
results
are
compared
and
aggregated
over
time
è several
hypotheses
è high
recall
• NER
based
on
MITIE
with
heuristics.