MediaEval Workshop and ACM Multimedia Conference

Author: Esra Acar

Recently I attended the MediaEval workshop and ACM Multimedia conference in Barcelona, Spain.


MediaEval is a multimedia benchmarking initiative that offers datasets and tasks to the community that emphasize the human and social aspects of multimedia.

Tasks offered this year at the MediaEval were:

  • Search and Hyperlinking of Television Content
  • Similar Segments in Social Speech
  • Placing: Geo-coordinate Prediction for Social Multimedia
  • Spoken Web Search: Spoken Term Detection for Low Resource Languages
  • Violent Scenes Detection in Film (Affect Task)
  • Social Event Detection for Social Multimedia
  • Visual Privacy: Preserving Privacy in Surveillance Videos
  • Emotion in Music (also an Affect Task)
  • Retrieving Diverse Social Images
  • MusiClef: Soundtrack Selection for Commercials
  • Crowdsourcing for Social Multimedia

The most interesting ones for me were Violent Scenes Detection in Film and Visual Privacy: Preserving Privacy in Surveillance Videos tasks.

We participated to the Visual Privacy task and won the Ghost in the Machine Award with our paper titled as DAI at the MediaEval 2013 Visual Privacy Task: Representing People with Foreground Edges!


The most interesting papers for me were:

The Vireo Team at MediaEval 2013: Violent Scenes Detection by Mid-level Concepts Learnt from Youtube

LIG at MediaEval 2013 Affect Task: Use of a Generic Method and Joint Audio-Visual Words

MediaEval 2013 Visual Privacy Task: Using Adaptive Edge Detection for Privacy in Surveillance Videos

You can find the MediaEval proceedings at

ACM Multimedia

ACM Multimedia is the premier conference and worldwide event bringing together multimedia experts and practitioners across academia and industry.

I presented our paper on violence detection in movies titled as Violence Detection in Hollywood Movies by the Fusion of Visual and Mid-level Audio Cues. I also participated to the Doctoral Symposium with my paper Learning Representations for Affective Video Understanding and got valuable feedback from participants!

There were two keynotes during the conference. The first keynote was given by Dr. Elizabeth F. Churchill from eBay Research Labs. She basically discussed multimedia from a different perspective and talked about how people experience multimedia. On the second day of the conference, Prof. Leonidas J. Guibas from Stanford University gave a keynote titled as The Space Between The Images. He mainly discussed expressing the relationships between different multimedia data items and talked about mathematical and algorithmic issues to compute relationships or mappings between media items.

Here are some papers that caught my attention:

  • Wow! You Are So Beautiful Today! (Best Paper Award)
  • Attributes-augmented Semantic Hierarchy for Image Retrieval (Best Student Paper Award)
  • Large-Scale Visual Sentiment Ontology and Detectors Using Adjective Noun Pairs
  • User Interest and Social Influence Based Emotion Prediction for Individuals

Leave a Reply

You can use these HTML tags

<a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>