MediaEval 2019: Multimedia Research Challenges
Register now: Data releases are beginning
Last day for regular registration: 30 June
MediaEval (Benchmarking Initiative for Multimedia Evaluation) offers
shared-tasks to the multimedia research community involving images,
text, video, music and speech. The tasks address cutting-edge
multimedia challenges (multimedia mining, retrieval, analysis, access
and exploration) with a clear human or social aspect. Our larger aim
is to promote reproducible research that makes multimedia a positive
force for society.
Further information and the link for registration to participate in
the tasks is available at: http://multimediaeval.org/mediaeval2019
Short descriptions of the 2019 tasks are below.
The MediaEval 2019 Workshop will take place 27-29 October 2019 near
Nice, France. (The workshop is scheduled so that participants can
combine the workshop with attendance at ACM Multimedia
https://www.acmmm.org/2019 in one trip, if they wish.)
#Emotion and Theme recognition in music using Jamendo#
Recognize emotions and themes conveyed in music recordings
(large-scale data set).
#Eyes and Ears Together: Multimodal coreference resolution#
Analyze videos to predict bounding boxes corresponding to nouns and
pronouns in the videos’ speech transcripts.
#GameStory: Video Game Analytics Challenge#
Analyze game streams (including audio and video streams, commentaries,
game data and statistics, interaction traces, viewer-to-viewer
communication) to carry out synchronization and event detection.
#Insight for Wellbeing: Multimodal personal health lifelog data analysis#
Analyze lifelogs (lifelog images, user-contributed tags, sensor
readings, weather/pollution information) to automatically make
well-being related predictions.
#Medico Medical Multimedia#
Analyze a multimodal dataset (videos, analysis data, study participant
data) to make predictions related to sperm quality.
#Multimedia Recommender Systems#
Participants can choose between one of two tasks that investigate the
use of multimedia content for recommendation.
#Multimedia Satellite Task: Flood Severity Estimation#
Analyze news reports (images/text) and/or satellite images for
information important for disaster management.
#No-audio Multimodal Speech Detection#
Participants receive videos (top view) and sensor readings
(acceleration and proximity) of people having conversations in a
natural social setting and are required to detect speaking turns.
Participants receive a set of images and are required to enhance them
in a way that blocks automatic inference of sensitive information,
while preserving image appeal. See: https://youtu.be/zShHPVOA070.
#Predicting Media Memorability#
Given a data set of multimedia content (images and/or videos) and
associated memorability annotations, automatically train a system to
#Scene Change (Brave New Task)#
Automatically create fun faux photo’s, composite images that fool you
at first, but can be identified as an imitation on closer inspection.
#Sports Video Annotation: Detection of Strokes in Table Tennis (Brave New Task)#
Automatically classify strokes in videos of table tennis.
#NewsFire: Discovering the triggers for viral news stories (Task force)#
Participants receive a large corpus of news stories and social media
posts (text and images) and are required to build a system that
detects the original triggers of news that spread with a viral or
For more information on the mission of MediaEval check out the videos
and proceedings from previous workshops, e.g.,
If you have further questions, please contact: Martha Larson
[log in to unmask]
On behalf of the MediaEval Community Council,
Prof. Bogdan IONESCU
ETTI - University Politehnica of Bucharest