[visionlist] OMG- Emotion Recognition Challenge - Final Results

[Pablo Barros]

Dear all,

I would like to share with you all the results of our first OMG- Emotion
Recognition Challenge.

Our challenge is based on the  One-Minute-Gradual Emotion Dataset
(OMG-Emotion Dataset), which is composed of 567  emotion videos with an
average length of 1 minute, collected from a variety of Youtube
channels. Each team had a task to describe each video with a continuous
space of arousal/valence domain.

The challenge had a total of 34 teams registered, from which we got 11
final submissions. Each final submission was composed of a  short paper
describing the solution and the link to the code repository.
The solutions used different modalities (ranging from unimodal audio and
vision to multimodal audio, vision, and text), and thus provide us with
a very complex evaluation scenario. All the submissions were based on
neural network models.

We split results into arousal and valence. For arousal, the best results
came from the GammaLab team. Their three submissions are our top 3 CCC
arousal, followed by the three submissions from the audEERING team, and
the two submissions from the HKUST-NISL2018
team.

For valence,  the GammaLab team stays still in first (with their three
submissions), followed by the two submissions of ADSC team and the three
submissions from the iBug team.

Congratulations to you all!

We provide a leaderboard on our website (
https://www2.informatik.uni-hamburg.de/wtm/OMG-EmotionChallenge/ ),
which will be permanently stored. This way, everyone can see the final
results of the challenge, have a quick access to a formal description of
the solutions and to the codes. This will help to disseminate knowledge
even further and will improve the reproducibility of your solutions.

We also provide a general leaderboard which will be updated constantly
with new submissions. If you are interested in having your score in our
general leaderboard, just send us an e-mail following the instructions
on our website.

I would also to invite you all to the presentation of the challenge
summary during the WCCI/IJCNN 2018 in Rio de Janeiro, Brasil.

Best Regards,

Pablo

-- 
Dr. Pablo Barros
Postdoctoral Research Associate - Crossmodal Learning Project (CML)
Knowledge Technology
Department of Informatics
University of Hamburg
Vogt-Koelln-Str. 30
22527 Hamburg, Germany
Phone: +49 40 42883 2535
Fax: +49 40 42883 2515
barros at informatik.uni-hamburg.de
https://www.inf.uni-hamburg.de/en/inst/ab/wtm/people/barros.html
https://www.inf.uni-hamburg.de/en/inst/ab/wtm/