Report from ICMR 2017

ACM International Conference on Multimedia Retrieval (ICMR) 2017

ACM ICMR 2017 in “Little Paris”

ACM ICMR is the premier International Conference on Multimedia Retrieval, and from 2011 it “illuminates the state of the arts in multimedia retrieval”. This year, ICMR was in an wonderful location: Bucharest, Romania also known as “Little Paris”. Every year at ICMR I learn something new. And here is what I learnt this year.

ICMR2017

Final Conference Shot at UP Bucharest

UNDERSTANDING THE TANGIBLE: object, scenes, semantic categories – everything we can see.

1) Objects (and YODA) can be easily tracked in videos.

Arnold Smeulders delivered a brilliant keynote on “things” retrieval: given an object in an image, can we find (and retrieve) it in other images, videos, and beyond? Very interesting technique for tracking objects (e.g. Yoda) in videos based on similarity learnt through siamese networks.

Tracking Yoda with Siamese Networks

2) Wearables + computer vision help explore cultural heritage sites.

As showed in his keynote, at MICC University of Florence, Alberto del Bimbo and his amazing team have designed smart audio guides for indoor and outdoor spaces. The system detects, recognises, and describes landmarks and artworks from wearable camera inputs (and GPS coordinates, in case of outdoor spaces).

3) We can finally quantify how much images provide complementary semantics compared to text [BEST MULTIMODAL PAPER AWARD].

For ages, the community has asked how relevant different modalities are for multimedia analysis: this paper (http://dl.acm.org/citation.cfm?id=3078991) finally proposes a solution to quantify information gaps between different modalities.

4) Exploring news corpuses is now very easy: news graphs are easy to navigate and aware of the type of relations between articles.

Remi Bois and his colleagues presented this framework (http://dl.acm.org/citation.cfm?id=3079023), made for professional journalists and the general public, for seamlessly browsing through large-scale news corpus. They built a graph where nodes are articles in a news corpus. The most relevant items to each article are chosen (and linked) based on an adaptive nearest neighbor technique. Each link is then characterised according to the type of relation of the 2 linked nodes.

5) Panorama outdoor images are much easier to localise.

In his beautiful work (https://t.co/3PHCZIrA4N), Ahmet Iscen from Inria developed an algorithm for location prediction from StreetView images, outperforming the state of the art thanks to an intelligent stitching pre-processing step: predicting locations from panoramas (stitched individual views) instead of individual street images improves performances dramatically!

UNDERSTANDING THE INTANGIBLE: artistic aspects, beauty, intent: everything we can perceive

1) Image search intent can be predicted by the way we look.

In his best paper candidate research work (http://dl.acm.org/citation.cfm?id=3078995), Mohammad Soleymani showed that image search intent (seeking information, finding content, or re-finding content) can be predicted from physisological responses (eye gaze) and implicit user interaction (mouse movements).

2) Real-time detection of fake tweets is now possible using user and textual cues.

Another best paper candidate (http://dl.acm.org/citation.cfm?id=3078979), this time from CERTH. The team collected a large dataset of fake/real sample tweets spanning 17 events and built an effective model from misleading content detection from tweet content and user characteristics. A live demo here: http://reveal-mklab.iti.gr/reveal/fake/

3) Music tracks have different functions in our daily lives.

Researchers from TU Delft have developed an algorithm (http://dl.acm.org/citation.cfm?id=3078997) which classifies music tracks according to their purpose in our daily activities: relax, study and workout.

4) By transferring image style we can make images more memorable!

The team at University of Trento built an automatic framework (https://arxiv.org/abs/1704.01745) to improve image memorability. A selector finds the style seeds (i.e. abstract paintings) which are likely to increase memorability of a given image, and after style transfer, the image will be more memorable!

5) Neural networks can help retrieve and discover child book illustrations.

In this amazing work (https://arxiv.org/pdf/1704.03057.pdf), motivated by real children experiences, Pinar and her team from Hacettepe University collected a large dataset of children book illustrations and found that neural networks can predict and transfer style, allowing to make “Winnie the witch”-like many other illustrations.

Winnie the Witch

6) Locals perceive their neighborhood as less interesting, more dangerous and dirtier compared to non-locals.

In this wonderful work (http://www.idiap.ch/~gatica/publications/SantaniRuizGatica-icmr17.pdf), presented by Darshan Santain from IDIAP, researchers asked locals and crowd-workers to look at pictures from various neighborhoods in Guanajuato and rate them according to interestingness, cleanliness, and safety.

THE FUTURE: What’s Next?

1) We will be able to anonymize images of outdoor spaces thanks to Instagram filters, as proposed by this work (http://dl.acm.org/citation.cfm?id=3080543) in the Brave New Idea session.  When an image of an outdoor space is manipulated with appropriate Instagram filters, the location of the image can be masked from vision-based geolocation classifiers.

2) Soon we will be able to embed watermarks in our Deep Neural Network models in order to protect our intellectual property [BEST PAPER AWARD]. This is a disruptive, novel idea, and that is why this work from KDDI Research and Japan National Institute of Informatics won the best paper award. Congratulations!

3) Given an image view of an object, we will predict the other side of things (from Smeulders’ keynote). In the pic: predicting the other side of chairs. Beautiful.

Predicting the other side of things

THANKS: To the organisers, to the volunteers, and to all the authors for their beautiful work 🙂

EDITORIAL NOTE: A more extensive report from ICMR 2017 by Miriam is available on Medium

Awarding the Best Social Media Reporters

The SIGMM Records team has adopted a new strategy to encourage the publication of information, and thus increase the chances to reach the community, increase knowledge and foster interaction. It consists of awarding the best Social Media reporters for each SIGMM conference, being the award a free registration to one of the SIGMM conference within a period of one year. All SIGMM members are welcome to participate and contribute, and are candidates to receive the award.

The Social Media Editors will issue a new open Call for Reports (CfR) via the Social Media channels every time a new SIGMM conference takes place, so the community can remember or be aware of this initiative, as well as can refresh its requirements and criteria.

The CfR will encourage activity on Social Media channels, posting information and contents related to the SIGMM conferences, with the proper hashtags (see our Recommendations). The reporters will be encouraged to mainly use Twitter, but other channels and innovative forms or trends of dissemination will be very welcome!

The Social Media Editors will be the jury for deciding the best reports (i.e., collection of posts) on Social Media channels, and thus will not qualify for this award. The awarded reporters will be additionally asked to provide a post-summary of the conference. The number of awards for each SIGMM conference is indicated in the table below. The awarded reporter will get a free registration to one of the SIGMM conferences (of his/her choice) within a period of one year.

Read more

Posting about SIGMM on Social Media

In Social Media, a common and effective mechanism to associate the publications about a specific thread, topic or event is to use hashtags. Therefore, the Social Media Editors believe in the convenience of recommending standards or basic rules for the creation and usage of the hashtags to be included in the publications related to the SIGMM conferences.

In this context, a common doubt is whether to include the ACM word and the year in the hashtags for conferences. Regarding the year, our recommendation is to not include it, as the date is available for the publications themselves and, this way, a single hashtag can be used to gather all the publications for all the editions of a specific SIGMM conference. Regarding the ACM word, our recommendation is to include it in the hashtag only if the conference acronym contains less than four letters (i.e., #acmmm, #acmtvx) and otherwise not (i.e., #mmsys, #icmr). Although consistency is important, not including ACM for MM (and for TVX) is clearly not a good identifier, and including it for MMSYS and ICMR results in a too long hashtag. Indeed, the #acmmmsys and #acmicmr hashtags have not been used before, contrarily to the wide use of #acmmm (and also of #acmtvx). Therefore, our recommendations for the usage and inclusion of hashtags can be summarized as:

Conference Hashtag

Include #ACM and #SIGMM?

MM #acmmm Yes
MMSYS #mmsys Yes
ICMR #icmr Yes

 

 

@sigmm on #SocialMedia

The new SIGMM Records team aims to extend the reach of relevant SIGMM-related news and events. It will also provide forums to stimulate discussion, interaction and collaboration between members of our community

The use of Social Media will be key to achieving the targeted mission. Initially, Twitter and Facebook will be used as the main Social Media channels for SIGMM. Youtube and LinkedIn will be used in a later stage.

Twitter (@sigmm) will be the main social media channel, for publishing information of interest in a variety of formats.

Facebook. It will include a Facebook page, ACM SIGMM, which will be used in a very similar manner than the Twitter account. In addition, a Facebook group will be created to have an interaction, discussion and collaboration forum.

The SIGMM and SIGMM Records websites will include Social Media icons, so the audience can share the contents on them via their personal Social Media channels.

Through the SIGMM Social Media and the personal communication channels, the Editors will encourage the community to contribute with interesting and relevant contents to be disseminated (e.g., outstanding contributions, Summer Schools, open positions, etc.).

The SIGMM Social Media channels will be particularly active during SIGMM sponsored events.

To promote community interaction, we recommend some policies for the creation and usage of the hashtags to be included in the publications related to the SIGMM conferences. They are summarized in the table below and can be accessed at this link. This will help the editors track contributions from the community and understand its impact.

 

Conference Hashtag

Include #ACM and #SIGMM?

MM #acmmm Yes
MMSYS #mmsys Yes
ICMR #icmr Yes

Apart from the SIGMM channels, many members of the community will contribute to publishing/sharing information of interest through their personal accounts (and ideally for their institutions’ ones), acting as Social Media reporters/advocates. The team includes: Christian Timmerer, Miriam Redi, Gwendal Simon, Michael Riegler, Wei Tsang Ooi, D. Ayman Shamma, and many others. The list is expected to grow, so please drop us a line if you are interested in joining us! Everybody is welcome to participate and contribute.

A further strategy to encourage the publication of information, and thus increase the chances to reach the community, increase knowledge and foster interaction, will consist of awarding those SIGMM members who provide the best posts and reports on Social Media for each SIGMM conference. The award will consist of a free registration to the next edition of the conference, and any SIGMM member is a candidate to get it. The awardees will be asked to provide a post-summary of the conference, which will be published on SIGMM Records. More details about the number of awards, their requirements and criteria can be found at this link.

We look forward to seeing you in our #SIGMM community! Follow us! 😉

photo_nial_murrayDr. Niall Murray (www.niallmurray.info) is a lecturer and researcher with the Faculty of Engineering and Informatics and the Software Research Institute in the Athlone Institute of Technology (AIT), Ireland. He received his BE (Electronic and Computer Engineering) from National University of Ireland, Galway (2003), MEng (Computer and Communication Systems) from the University of Limerick (2004) and PhD in 2014. Since 2004, he has worked in R&D roles across a number of industries: telecommunications, finance, health and education. In 2014 he founded the Truly Immersive and Interactive Multimedia Experiences lab (TIIMEx). His research interests include immersive multimedia communication, multimedia synchronization, multisensory multimedia, quality of experience (QoE) and wearable sensor systems. In this context, TIIMEx builds and evaluates from a user perceived quality perspective, end-to-end communication systems and novel immersive and interactive applications.

 

 

xavier-giróXavier Giro-i-Nieto is an associate professor at the Universitat Politecnica de Catalunya (UPC). He graduated in Telecommuncations Engineering studies at ETSETB (UPC) in 2000, after completing his master thesis on image compression at the Vrije Universiteit in Brussels (VUB). He obtained his Phd on image retrieval in 2012, under the supervision by Professor Ferran Marqués from UPC and Professor Shih-Fu Chang from Columbia University. He was a visiting scholar during Summers 2008 to 2014 at the Digital Video and MultiMedia laboratory at Columbia University, in New York. He has served as area chair in ACM Multimedia 2016 and is currently a member of the editorial board of IEEE Transactions on Multimedia. His current research interests are focus on applying deep learning to multimodal applications, such as video analytics, eye gaze prediction and lifelogging.

 

 

LexingXieLexing Xie is Associate Professor in the Research School of Computer Science at the Australian National University, she leads the ANU Computational Media lab (http://cm.cecs.anu.edu.au). Her research interests are in machine learning, multimedia, social media. Of particular recent interest are stochastic time series models, neural network for sequences, and active learning, applied to diverse problems such as multimedia knowledge graphs, modeling popularity in social media, joint optimization and structured prediction problems, and social recommendation. Her research is supported from the US Air Force Office of Scientific Research, Data61, Data to Decisions CRC and the Australian Research Council. Lexing’s research has received six best student paper and best paper awards in ACM and IEEE conferences between 2002 and 2015. She is IEEE Circuits and Systems Society Distinguished Lecturer 2016-2017. She currently serves an associate editor of ACM Trans. MM, ACM TiiS and PeerJ Computer Science. Her service roles include the program and organizing committees of major multimedia, machine learning, web and social media conferences. She was research staff member at IBM T.J. Watson Research Center in New York from 2005 to 2010.

photo_mario_montagudDr. Mario Montagud (@mario_montagud) was born in Montitxelvo (Spain). He received a BsC in Telecommunications Engineering in 2011, an MsC degree in “Telecommunication Technologies, Systems and Networks” in 2012 and a PhD degree in Telecommunications (Cum Laude Distinction) in 2015, all of them at the Polytechnic University of Valencia (UPV). During his PhD degree and after completing it, he did 3 research stays (accumulating 18 months) at CWI (The National Research Institute for Mathematics and Computer Science in the Netherlands). He also has experience as a postdoc researcher at UPV. His topics of interest include Computer Networks, Interactive and Immersive Media, Synchronization, and QoE (Quality of Experience). Mario is (co-) author of over 50 scientific and teaching publications, and has contributed to standardization within the IETF (Internet Engineering Task Force). He is member of the Technical Committee of several international conferences (e.g., ACM MM, MMSYS and TVX), co-organizer of the international MediaSync Workshop series, and member of the Editorial Board of international journals. He is also lead editor of “MediaSync: Handbook on Multimedia Synchronization” (Springer, 2017) and Communication Embassador of ACM SIGCHI (Special Interest Group on Computer-Human Interaction). Webpage: https://sites.google.com/site/mamontor/

Introduction to the Opinion Column

Welcome to the SIGMM Community Discussion Column! In this very first edition we would like to introduce the column to the community, its objectives and main operative characteristics.

Given the exponential amount of multimedia data shared online and offline everyday, research in Multimedia is of unprecedented importance. We might be now facing a new era of our research field, and we would like the whole community to be involved in the improvement and evolution of our domain.

The column has two main goals. First, we will promote dialogue regarding topics of interests for the MM community, by providing tools for continuous discussion among the members of the multimedia community. Every quarter, we will discuss (usually) one topic via online tools. Topics will include “What is Multimedia, and what is the role of the Multimedia community in science?”; “Diversity and minorities in the community”; “The ACM code of ethics”; etc.

Second, we will monitor and summarize on-going discussions, and spread their results within and outside the community. Every edition of this column will then summarize the discussion, highlighting popular and non-popular opinions, agreed action points and future work.

To foster the discussion, we set up an online discussion forum to which all members of the multimedia community (expertise and seniority mixed) can participate: the Facebook MM Community Discussion group (follow this link: https://www.facebook.com/groups/132278853988735/) . For every edition of the column, we will choose an initial set of topics of high relevance for the community. We will include, for example, topics that have been previously discussed at ACM meetings (e.g., the code of ethics), or in related events (e.g., Diversity at MM Women lunch), or popular off-line discussions among MM researchers (e.g., review processes, vision of the scientific community…). In the first 15 days of the quarter, the members of the community will choose one topic from this short-list via an online poll shared through the MM Facebook group. We will then select the topic that received the higher number of votes as the subject for the quarterly discussion.

Volunteers or selected members of the MM group will start the discussion via Facebook posts on the group page. The discussion will be then open for a period of a month. All members of the community can participate by replying to posts or by directly posting on the group page, describing their point of view on the subject while being concise and clear. During this period, we will monitor and moderate (when needed) the discussion. At the end of the month, we will summarise the discussion by describing its evolution, exposing major and minor opinions, outlining highlights and lowlights. A final text with the summary and some relevant discussion extracts will be prepared and will appear in the SIGMM Records and in the Facebook “MM Community page”: https://www.facebook.com/MM-Community-217668705388738/.

Hopefully, the community will benefit from this initiative by either reaching some consensus or by pointing out important topics that are not mature enough and require further exploration. In the long-term, we hope these process will make the community evolve through large consensus and bottom-up discussions.

Let’s contribute and foster research around topics of high interest for the community!

Xavi and Miriam

Xavier Almeda-PinedaDr. Xavier Alameda-Pineda (Xavi) is research scientist at INRIA. Xavi’s interdisciplinary background (Msc in Mathematics, Telecommunications and Computer Science) grounded him to pursue his PhD in Mathematics and Computer Science, and a further post-doc in the University of Trento. His research interests are signal processing, computer vision and machine learning for scene and behavior understanding using multimodal data. He is the winner of the best paper award of ACM MM 2015, the best student paper award at IEEE WASPAA 2015 and the best scientific paper award at IAPR ICPR 2016.

 

 

 

Mariam RediDr. Miriam Redi is a research scientist in the Social Dynamics team at Bell Labs Cambridge. Her research focuses on content-based social multimedia understanding and culture analytics. In particular, Miriam explores ways to automatically assess visual aesthetics, sentiment, and creativity and exploit the power of computer vision in the context of web, social media, and online communities. Previously, she was a postdoc in the Social Media group at Yahoo Labs Barcelona and a research scientist at Yahoo London. Miriam holds a PhD from the Multimedia group in EURECOM, Sophia Antipolis.

@sigmm Records: serving the community

The SIGMM Records are renewing, with the continued ambition of being a useful resource for the multimedia community. We want to provide a forum for (open) discussion, but also to become the primary source of information for our community.

Firstly, I would like to thank Carsten who was run, single-handed, the whole records for many many years. We all agree that he has done an amazing job, and that his service deserves our gratitude, and possibly some beers, when you meet him at conferences and meetings.

As you are probably aware, a number of changes in the records are underway. We want your opinions and suggestions to make this resource the best it can be. Hence, we need your help to make this a success, so please drop us a line if you want to join the team.

The two main visible changes are:

We have a new amazing team to lead the records in the coming years. I am so glad to have their help: http://sigmm.hosting.acm.org/impressum/

We have reorganized the records and its structure, in three main clusters:

More changes to come. Stay tuned!

Pablo (Editor in Chief) + Carsten and Mario (Information Directors)

Pablo CesarDr. Pablo Cesar leads the Distributed and Interactive Systems group at Centrum Wiskunde & Informatica (CWI) in the Netherlands. Pablo’s research focuses on modeling and controlling complex collections of media objects (including real-time media and sensor data) that are distributed in time and space. His fundamental interest is in understanding how different customizations of such collections affect the user experience. Pablo is the PI of Public Private Partnership projects with Xinhuanet and ByBorre, and very successful EU-funded projects like 2-IMMERSE, REVERIE and Vconect. He has (co)-authored over 100 articles. He is member of the editorial board of, among others, ACM Transactions on Multimedia (TOMM). Pablo has given tutorials about multimedia systems in prestigious conferences such as ACM Multimedia, CHI, and the WWW conference. He acted as an invited expert at the European Commission’s Future Media Internet Architecture Think Tank and participates in standardisation activities at MPEG (point-cloud compression) and ITU (QoE for multi-party tele-meetings). Webpage: http://homepages.cwi.nl/~garcia/

 

Carsten GriwodzDr. Carsten Griwodz is Chief Research Scientist at the Media Department of theNorwegian research company Simula Research Laboratory AS, Norway, and professor at the University of Oslo. He is also co-founder of ForzaSys AS, a social media startup for sports. He is steering committee member of ACM MMSys and ACM/IEEE NetGames. He is associate editor of the IEEE MMTC R-Letter and was previously editor-in-chief of the ACM SIGMM Records and editor of ACM TOMM.

 

 

photo_mario_montagudDr. Mario Montagud (@mario_montagud) was born in Montitxelvo (Spain). He received a BsC in Telecommunications Engineering in 2011, an MsC degree in “Telecommunication Technologies, Systems and Networks” in 2012 and a PhD degree in Telecommunications (Cum Laude Distinction) in 2015, all of them at the Polytechnic University of Valencia (UPV). During his PhD degree and after completing it, he did 3 research stays (accumulating 18 months) at CWI (The National Research Institute for Mathematics and Computer Science in the Netherlands). He also has experience as a postdoc researcher at UPV. His topics of interest include Computer Networks, Interactive and Immersive Media, Synchronization, and QoE (Quality of Experience). Mario is (co-) author of over 50 scientific and teaching publications, and has contributed to standardization within the IETF (Internet Engineering Task Force). He is member of the Technical Committee of several international conferences (e.g., ACM MM, MMSYS and TVX), co-organizer of the international MediaSync Workshop series, and member of the Editorial Board of international journals. He is also lead editor of “MediaSync: Handbook on Multimedia Synchronization” (Springer, 2017) and Communication Embassador of ACM SIGCHI (Special Interest Group on Computer-Human Interaction). Webpage: https://sites.google.com/site/mamontor/