Tuesday, July 30, 2013

MMM 2014 - The 20th Anniversary International Conference on MultiMedia Modeling Dublin, Ireland

Now in its 20th year, MMM is a leading international conference for researchers and industry practitioners for sharing new ideas, original research results and practical development experiences from all MMM related areas. The conference calls for full papers and short papersreporting original research and investigation results as well as demonstrations in all areas related to multimedia modeling technologies and applications. There are four special sessions (details below) and MMM2014 features the 2014 Video Browser Showdown.

Submissions should conform to the formatting instructions of Springer Verlag, LNCS series. To submit to MMM 2014, please go to the MMM 2014 submission site.

Special Sessions papers are called for for any of these five special sessions:

  • Social Geo-Media Analytics and Retrieval
  • Multimedia Hyperlinking and Retrieval
  • 3D Multimedia Computing and Modeling
  • Multimedia Analysis for Surveillance Video and Security
  • Mediadrom: artful post-TV scenarios

The conference proceedings will be published as series of Lecture Notes in Computer Science (LNCS) by Springer. Authors of selected papers of MMM 2014 will be invited to publish extended versions in a special issue of Multimedia Tools and Applications journal (MTAP, Springer).


The topics of interest for MMM 2014 include, but are not limited to:

Multimedia Content Analysis

  • Multimedia Indexing
  • Multimedia Mining
  • Multimedia Abstraction and Summarization
  • Multimedia Annotation, Tagging and Recommendation
  • Multimodal Analysis for Retrieval Applications
  • Semantic Analysis of Multimedia and Contextual Data
  • Multimedia Fusion Methods
  • Media Content Browsing and Retrieval Tools

Multimedia Signal Processing and Communications

  • Media Representation and Algorithms
  • Audio, Image, Video Processing, Coding and Compression
  • Multimedia Security and Content Protection
  • Multimedia Standards and Related Issues
  • Advances in Multimedia Networking and Streaming
  • Multimedia Databases, Content Delivery and Transport
  • Wireless and Mobile Multimedia Networking

Multimedia Applications and Services

  • Multi-Camera and Multi-View Systems
  • Augmented and Virtual Reality, Virtual Environments
  • Real-Time and Interactive Multimedia Applications
  • Mobile Multimedia Applications
  • Multimedia Web Applications
  • Multimedia Authoring & Personalisation
  • Interactive Multimedia and Interfaces
  • Sensor Networks (Video Surveillance, Distributed Systems)
  • Emerging Trends (e-learning, e-Health, Social Media, Multimedia Collaboration, etc.)

Other topics related to MultiMedia Modeling but not listed above are welcome as well.

Submissions should conform to the formatting instructions of Springer Verlag, LNCS series All long and short paper submissions will be peer-reviewed in a double-blind review process to ensure maximum quality. Demonstration submissions do not need to be anonymised.

Important Dates:

  • Aug 09, 2013: Deadline for papers (full, short, sp. sessions) & demos
  • Sep 18, 2013:Acceptance notification for all paper types and demos
  • Oct 16, 2013: Camera-ready versions due


Vienna University of Technology, Austria
Do you want to be an expert on information retrieval - the technology battle field between IT giants? Do you want to understand what it means to answer a user's request for information? What it means to answer it _well_? Do you have the technical skills to develop applications to run in the cloud and the mathematical understanding to interpret the significance of your own system's results? If so, you are invited to read below and contact us in the city with the highest quality of living in the world [1].
The Information and Software Engineering Group at the Institute of Software Technology and Interactive Systems at the Vienna University of Technology announces the availability of a 3-year PhD position.
The position is associated to the ADmIRE project (Abstracting Domain-Specific Information Retrieval and Evaluation), funded by the Austrian Science Fund (FWF). The project will start in October 2013. Its main objective is to provide the foundation for the practice of Evidence-based IR, through creating a framework and tools for systematic reviews of IR results, and developing a methodology for conducting component-level evaluation of IR systems based on a workflow paradigm.
The main task of the PhD student is to develop a methodology for conducting systematic reviews of IR experimental results (together with the PostDocs working on the project) and to develop and implement tools to support systematic reviews by performing search for IR papers relevant to a research question, and semi-automated analysis of the texts of relevant IR papers and their results. Systematic reviews aim to synthesise the evidence supporting answers to a research question by systematically analysing the literature and performing a quantitative synthesis and meta-analysis of the experimental results of relevant papers. Systematic reviews are widely used in medicine, and are being adopted in many domains with a strong empirical approach.
To approach this task, the PhD student must have strong mathematical background (preferably including statistics), as well as a fundamental understanding of Information Retrieval. Applicants with experience in NLP will be favoured.
The position is available from October 2013 and is open until filled. The position is fulltime (40 hours a week).
For further information, contact Applications for the position (including a CV) should also be sent to this e-mail address.
== About our Group ==
The working group Information and Software Engineering is part of the Institute of Software Technology and Interactive Systems. It covers the design and development of software and information systems in research and education. The other groups cover E-Commerce, Information Systems, and Interactive Media Systems. Both theoretical and engineering-like approaches to problem solving have equal balance in our research and educational work - leading to "real world problem solving".
The Group is situated in downtown Vienna, in an environment which brings together all aspects of student and academic life. The PhD student will be part of the Information Management & Preservation (IMP) Lab,  which consists of researchers in text, music and multimedia information retrieval, as well as in digital preservation and eScience. The working language is English. A more detailed description of the IMP lab is available at
[1] 2012 Quality of Living worldwide city rankings – Mercer survey

Thursday, July 18, 2013

[CVPR 2013] Three Trending Computer Vision Research Areas

As I walked through the large poster-filled hall at CVPR 2013, I asked myself, “Quo vadis Computer Vision?" (Where are you going, computer vision?)  I see lots of papers which exploit last year’s ideas, copious amounts of incremental research, and an overabundance of off-the-shelf computational techniques being recombined in seemingly novel ways.  When you are active in computer vision research for several years, it is not rare to find oneself becoming bored by a significant fraction of papers at research conferences.  Right after the main CVPR conference, I felt mentally drained and needed to get a breath of fresh air, so I spent several days checking out the sights in Oregon.  Here is one picture -- proof that the CVPR2013 had more to offer than ideas!

When I returned from sight-seeing, I took a more circumspect look at the field of computer vision.  I immediately noticed that vision research is actually advancing and growing in a healthy way.  (Unfortunately, most junior students have a hard determining which research papers are actually novel and/or significant.)  A handful of new research themes arise each year, and today I’d like to briefly discuss three new computer vision research themes which are likely to rise in popularity in the foreseeable future (2-5 years).

1) RGB-D input data is trending.

Many of this year’s papers take a single 2.5D RGB-D image as input and try to parse the image into its constituent objects.  The number of papers doing this with RGBD data is seemingly infinite.  Some other CVPR 2013 approaches don’t try to parse the image, but instead do something else like: fit cuboids, reason about affordances in 3D, or reason about illumination.  The reason why such inputs are becoming more popular is simple: RGB-D images can be obtained via cheap and readily available sensors such as Microsoft’s Kinect.  Depth measurements used to be obtained by expensive time of flight sensors (in the late 90s and early 00s), but as of 2013, $150 can buy you one these depth sensing bad-boys!  In fact, I had bought a Kinect just because I thought that it might come in handy one day -- and since I’ve joined MIT, I’ve been delving into the RGB-D reconstruction domain on my own.  It is just a matter of time until the newest iPhone has an on-board depth sensor, so the current line of research which relies on RGB-D input is likely to become the norm within a few years.

H. Jiang and J. Xiao. A Linear Approach to Matching Cuboids in RGBD Images. In CVPR 2013. [pdf] [code]

2) Mid-level patch discovery is a hot research topic.
Saurabh Singh from CMU introduced this idea in his seminal ECCV 2012 paper, and Carl Doersch applied this idea to large-scale Google Street-View imagery in the “What makes Paris look like Paris?” SIGGRAPH 2012 paper.  The idea is to automatically extract mid-level patches (which could be objects, object parts, or just chunks of stuff) from images with the constraint that those are the most informative patches.  Regarding the SIGGRAPH paper, see the video below.

Unsupervised Discovery of Mid-Level Discriminative PatchesSaurabh Singh, Abhinav Gupta, Alexei A. Efros. In ECCV, 2012.

Carl Doersch, Saurabh Singh, Abhinav Gupta, Josef Sivic, and Alexei A. Efros. What Makes Paris Look like Paris? In SIGGRAPH 2012. [pdf]
At CVPR 2013, it was evident that the idea of "learning mid-level parts for scenes" is being pursued by other top-tier computer vision research groups.  Here are some CVPR 2013 papers which capitalize on this idea:
Blocks that Shout: Distinctive Parts for Scene Classification. Mayank Juneja, Andrea Vedaldi, CV Jawahar, Andrew Zisserman. In CVPR, 2013. [pdf]
Representing Videos using Mid-level Discriminative Patches. Arpit Jain, Abhinav Gupta, Mikel Rodriguez, Larry Davis. CVPR, 2013. [pdf]
Part Discovery from Partial Correspondence. Subhransu Maji, Gregory Shakhnarovich. In CVPR, 2013. [pdf]
3) Deep-learning and feature learning are on the rise within the Computer Vision community.
It seems that everybody at Google Research is working on Deep-learning.  Will it solve all vision problems?  Is it the one computational ring to rule them all?  Personally, I doubt it, but the rising presence of deep learning is forcing every researcher to brush up on their l33t backprop skillz.  In other words, if you don't know who Geoff Hinton is, then you are in trouble.

PhD student in Computer Vision and Machine Learning

Faculty of Science - Informatics Institute

The Faculty of Science occupies a leading position internationally in its fields of research and participates in a large number of cooperative programs with universities, research institutes and businesses. The faculty has a student body of around 3,000 and 1,500 members of staff, spread over eight research institutes and a number of faculty-wide support services. A considerable part of the research is made possible by external funding from Dutch and international organizations and the private sector. The Faculty of Science offers thirteen Bachelor's degree programs and eighteen Master’s degree programs in the fields of the exact sciences, computer science and information studies, and life and earth sciences.

Since September 2010, the whole faculty has been housed in a brand new building at the Science Park in Amsterdam. The instalment of the faculty has made the Science Park one of the largest centers of academic research in the Netherlands.

The Informatics Institute (IvI) is one of the large research institutes with the faculty, with a focus on complex information systems divided in two broad themes: 'Computational Systems' and 'Intelligent Systems.' The institute has a prominent international standing and is active in a dynamic scientific area, with a strong innovative character and an extensive portfolio of externally funded projects.

Project description

The topic of the PhD is to recognize objects in a visual data stream. In such a stream the object classes of interest shift over time. Hence, the traditional approach to learn classifiers for a predefined set of objects is unsuited. A promising approach in classifying unseen objects into a novel category is to learn a semantic attribute image representation. The aim for this PhD is to develop new algorithms to learn such a high-level semantic representation from weakly annotated images and to learn the mapping to an unknown class from freely available (textual) sources. Another project aim is to model the visual data stream to understand which images or novel concepts could become a visual trend.

  • Master degree in Artificial Intelligence, Computer Science or related field;
  • excellent programming skills (the project is in Matlab, Python and C/C++);
  • solid mathematics foundations, especially statistics and linear algebra;
  • highly motivated;
  • fluent in English, both written and spoken;
  • proven experience with computer vision and/or machine learning is a big plus.
Further information

The position is within the Intelligent Systems Lab Amsterdam (ISLA) and will be supervised by dr. Thomas Mensink and dr. Cees Snoek. The position is part of a 5-year Personal VIDI Grant funded by the Dutch Organization for Scientific Research. The successful candidate will work in a stimulating environment of a leading and highly active research team including one faculty member, a post-doc and six PhD students. The team has repeatedly won the major visual search competitions, including NIST TRECVID, PASCAL Visual Object Challenge, ImageCLEF, and the ImageNet Large Scale Visual Recognition Challenge.

Informal inquiries on the position can be sent by email to:

You may also consult the following websites for more information:


Starting date: autumn 2013. The appointment will be on a temporary basis for a period of 4 years (initial appointment will be for a period of 18 months and after satisfactory evaluation it can be extended for a total duration of 4 years) and should lead to a dissertation (PhD thesis). An educational plan will be drafted that includes attendance of courses and (international) meetings. The PhD student is also expected to assist in teaching of undergraduates.

Based on a full-time appointment (38 hours per week) the gross monthly salary will range from €2,062 in the first year to €2,664 in the last year. There are also secondary benefits, such as 8% holiday allowance per year and the end of year allowance of 8.3%. The Collective Employment Agreement (CAO) of the Dutch Universities is applicable.

Some of the things we have to offer:

  • Competitive pay and excellent benefits;
  • extremely friendly, interactive and international working environment;
  • new building near the city center of one of Europe's most beautiful and lively cities;
  • direct access to high-end computing facilities.

English is the working language within the Informatics Institute. Moreover, since Amsterdam is a very international city where almost everybody speaks and understands English, candidates need not be afraid of the language barrier.

Job application

Applications may only be submitted by electronic mail by clicking the Apply now–button below or by sending your application to To process your application immediately, please quote the vacancy number 13-205 and the position you are applying for in the subject-line. Applications must include a motivation letter explaining why you are the right candidate, curriculum vitae, (max 2 pages), a copy of your master’s thesis, a list of projects you have worked on (with brief descriptions of your contributions, max 2 pages) and the names and contact addresses of two academic references. All these should be grouped in one PDF attachment.

LIRE 0.9.4 beta released

Article from

I’ve just uploaded LIRE 0.9.4 beta to the Google Code downloads page. This is an intermediate release that reflects several changes within the SVN trunk. Basically I put it online as there are many, many bugs solved in this one and it’s performing much, much faster than the 0.9.3 release. If you want to get the latest version I’d recommend to stick to the SVN. However, currently I’m changing a lot of feature serialization methods, so there’s no guarantee that an index created with 0.9.4 beta will work out with any newer version. Note also that the release does not work with older indexes ;)

Major changes include, but are not limited to:

  • New features: PHOG, local binary patterns and binary patterns pyramid
  • Parallel indexing: a producer-consumer based indexing application that makes heavy use of available CPU cores. On a current Intel Core i7 or considerably large Intel Xeon system it is able to reduce extraction to a marginal overhead to disk I/O.
  • Intermediate byte[] based feature data files: a new way to extract features in a distributed way
  • In-memory cached ImageSearcher: as long as there is enough memory all linear searching is done in memory without much disk I/O (cp. class GenericFastImageSearcher and set caching to true)
  • Approximate indexing based on hashing: tests with 1.5 million led to search time < 300ms (cp. GenericDocumentBuilder with hashing set to true and BitSamplingImageSearcher)
  • Footprint of many global descriptors has been significantly reduced. Examples: EdgeHistogram 40 bytes, ColorLayout 504 bytes, FCTH 96 bytes, …
  • New unit test for benchmarking features on the UCID data set.

All changes can be found in the CHANGES.txt file.

2 θέσεις Μηχανικών Υπολογιστών ή Επιστημόνων Πληροφορικής στην Ξάνθη

(1 θέση μεταδιδακτορικού ερευνητή και 1 θέση υποψήφιου διδάκτορα)

Στα πλαίσια δύο νέων ερευνητικών προγραμμάτων που έχουν εγκριθεί και πρόκειται να ξεκινήσουν άμεσα, ο Τομέας Λογισμικού της Πολυτεχνικής Σχολής Ξάνθης, ΔΠΘ, αναζητά 2 επιστημονικούς συνεργάτες (ένα μεταδιδακτορικό ερευνητή και ένα υποψήφιο διδάκτορα), για το “Συνεργαστήριο Μηχανικής Λογισμικού για το Περιβάλλον”. Αντικείμενο των θέσεων είναι να προάγουν την έρευνα στην γενικότερη περιοχή της οικοπληροφορικής, και ειδικότερα στις παρακάτω περιοχές:

  • Domain Specific Programming Languages
  • Semantic Web, Knowledge Modelling
  • Service Oriented Systems
  • Open Big Data and Data Interoperability
  • Intelligent Software Systems
  • Requirements Engineering

Ο μεταδιδακτορικός ερευνητής αναμένεται να έχει αναγνωρισμένη ερευνητική συνεισφορά σε κάποια από τις παραπάνω περιοχές.

Ο υποψήφιος διδάκτορας αναμένεται να έχει πολύ καλό υπόβαθρο στην επιστήμη των υπολογιστών και ενδιαφέρον για πρωτότυπη έρευνα στις παραπάνω περιοχές.

Οι ιδανικοί υποψήφιοι για τη θέση πέρα των τυπικών προσόντων, απαραίτητο είναι να έχουν άριστη γνώση της αγγλικής γλώσσας, εξαιρετική ευχέρεια στον προγραμματισμό υπολογιστών, ομαδικό πνεύμα και διάθεση για διαθεματική έρευνα. Σχετική εμπειρία σε εφαρμογές για το περιβάλλον, την οικολογία, την κλιματική αλλαγή ή τη γεωργία θα θεωρηθεί επιπλέον προσόν, αλλά δεν είναι απαραίτητη.

Οι θέσεις είναι πλήρους απασχόλησης, με διάρκεια ενός έτους με δυνατότητα ανανέωσης για ακόμη 2 χρόνια, και χρηματοδοτούνται από τα ερευνητικά προγράμματα:

  1. ALPINE (ΓΓΕΤ “Πρόγραμμα Συνεργασία”, 2013-2015)
    Ευφυής αρχιτεκτονική δικτύων αισθητήρων χαμηλής κατανάλωσης για περιβαλλοντική διαχείριση

    Σκοπός του έργου είναι ο σχεδιασμός, η υλοποίηση και η επίδειξη μιας ευφυούς αρχιτεκτονικής δικτύων αισθητήρων χαμηλής κατανάλωσης, μέσω μιας εργαλειοθήκης προγραμματισμού υψηλού επιπέδου, η οποία είναι προσαρμοσμένη στις ανάγκες περιβαλλοντικής διαχείρισης σε πραγματικό χρόνο. Η λύση που παρέχει το έργο είναι γενικής χρήσης, και επιδεικνύεται σε δύο πιλοτικές εφαρμογές. Η πρώτη αφορά τη διαχείριση υποδομών υδατικών πόρων και η δεύτερη την παρατήρηση και προστασία της άγριας ζωής. Η ερευνητική συνεισφορά της ομάδας μας είναι στην μοντελοποίηση της γνώσης στις δύο πιλοτικές εφαρμογές με τη μορφή οντολογιών, και στην ανάπτυξη μιας γλώσσας προγραμματισμού ειδικού πεδίου για τη διαχείριση και ολοκλήρωση πληροφοριών από δίκτυα περιβαλλοντικών αισθητήρων.

  2. MODEXTREME (Ευρωπαϊκό πρόγραμμα, FP7, 2013-2016)
    Modelling vegetation response to extreme events

    Το έργο MODEXTREME στοχεύει να βοηθήσει την Ευρωπαϊκή και παγκόσμια γεωργία να αντιμετωπίσει τα ακραία κλιματικά φαινόμενα μέσα από τη βελτίωση των ικανοτήτων των βιοφυσικών μοντέλων ανάπτυξης φυτών να προσομοιώνουν αγροτική παραγωγή σε συνθήκες κλιματικής αλλαγής. Η ομάδα μας συνεισφέρει (α) στο σχεδιασμό και την ανάπτυξη βιβλιοθηκών λογισμικού που υλοποιούν μοντέλα προσομοίωσης ανάπτυξης φυτών, και (β) τη διαλειτουργικότητα υπηρεσιών και δεδομένων σε μια πλατφόρμα προσομοίωσης και σύγκρισης πολλών μοντέλων ανάπτυξης φυτών.

Για πληροφορίες σχετικά με την ερευνητική μας ομάδα δείτε:

Για εκδήλωση ενδιαφέροντος στείλτε το βιογραφικό σας και ένα συνοδευτικό email στον Γιάννη Αθανασιάδη, επίκουρο καθηγητή, Τμήμα Ηλεκτρολόγων Μηχανικών και Μηχανικών Υπολογιστών, ΔΠΘ, στην διεύθυνση με τίτλο jobs-2013, μέχρι τις 20 Αυγούστου 2013.