Thursday, January 26, 2012

Mobile Visual Search (MVS)

An amazing presentation by Oge Marques!!!!

Watch the (full length !!! ) video here:



Mobile Visual Search (MVS) is a fascinating research field with many open challenges and opportunities which have the potential to impact the way we organize, annotate, and retrieve visual data (images and videos) using mobile devices. This talk is structured in four parts:

(i) MVS — opportunities: where I present recent and relevant numbers of the mobile computing market, particularly in the field of photography apps, social networks, and mobile search.

(ii) Basic concepts: where I explain the basic MVS pipeline and discuss the three main MVS scenarios and associated challenges.

(iii) Advanced technical details: where I explain technical aspects of feature extraction, indexing, descriptor matching, and geometric verification, discuss the state of the art in these fields, and comment on open problems and research opportunities.

(iv) Examples and applications: where I show recent and significant examples of academic research (e.g., Stanford Product Search System) and commercial apps (e.g., Google Goggles, oMoby, kooaba) in this field.

Great stuff professor Marques, thanks for sharing

Wednesday, January 25, 2012


Στα πλαίσια του ερευνητικού προγράμματος ΘΑΛΗΣ, η Ομάδα Επεξεργασίας Εικόνας και Πολυμέσων ( του Τμήματος Ηλεκτρολόγων Μηχανικών και Μηχανικών Υπολογιστών του Δημοκριτείου Πανεπιστημίου Θράκης, θα διερευνήσει νέες μεθοδολογίες για την αποτελεσματική ανάκτηση τριδιάστατων (3Δ) αντικειμένων (στατικών και 3Δ βίντεο).

Σε αυτό το πρόγραμμα υπάρχουν δύο (2) χρηματοδοτούμενες θέσεις για εκπόνηση διδακτορικών διατριβών.

Οι θέσεις είναι άμεσα διαθέσιμες. Παρόλη την άμεση διαθεσιμότητα, ενθαρρύνονται να εκδηλώσουν το ενδιαφέρον τους υποψήφιοι οι οποίοι διαθέτουν τα απαιτούμενα προσόντα και θα έχουν ολοκληρώσει τις βασικές ή μεταπτυχιακές σπουδές τους μέχρι τον ερχόμενο Σεπτέμβρη.

Είναι επιθυμητό οι υποψήφιοι να έχουν υπόβαθρο σε τουλάχιστον κάποιο από τα θέματα που αφορούν σε Γραφικά, Επεξεργασία Εικόνας και Αναγνώριση Προτύπων καθώς και να έχουν επαρκή εμπειρία στον προγραμματισμό είτε σε περιβάλλον Matlab είτε σε οποιοδήποτε άλλο περιβάλλον προγραμματισμού C/C++/C#.

Οι υποψήφιοι που θα επιλεγούν θα δουλέψουν σε ένα δυναμικό περιβάλλον και θα συνεργασθούν με ερευνητές ιδρυμάτων του εσωτερικού (Εθνικό Καποδιστριακό Πανεπιστήμιο Αθηνών – Τμήμα Πληροφορικής, Ε.Κ. «ΑΘΗΝΑ» – Ινστιτούτο Πολιτιστικής και Εκπαιδευτικής Τεχνολογίας, ΕΚΕΦΕ «ΔΗΜΟΚΡΙΤΟΣ» - Ινστιτούτο Πληροφορικής και Τηλεπικοινωνιών) καθώς και με ερευνητές ιδρυμάτων του εξωτερικού (University of Houston – USA, Vrije Universiteit Brussel – Belgium, Utrecht University - Netherlands, Consiglio Nazionale delle Ricerche - Italy).

Friday, January 20, 2012

The faster-than-fast Fourier transform

Article from MITNews

For a large range of practically useful cases, MIT researchers find a way to increase the speed of one of the most important algorithms in the information sciences.

The Fourier transform is one of the most fundamental concepts in the information sciences. It’s a method for representing an irregular signal — such as the voltage fluctuations in the wire that connects an MP3 player to a loudspeaker — as a combination of pure frequencies. It’s universal in signal processing, but it can also be used to compress image and audio files, solve differential equations and price stock options, among other things.
The reason the Fourier transform is so prevalent is an algorithm called the fast Fourier transform (FFT), devised in the mid-1960s, which made it practical to calculate Fourier transforms on the fly. Ever since the FFT was proposed, however, people have wondered whether an even faster algorithm could be found.
At the Association for Computing Machinery’s Symposium on Discrete Algorithms (SODA) this week, a group of MIT researchers will present a new algorithm that, in a large range of practically important cases, improves on the fast Fourier transform. Under some circumstances, the improvement can be dramatic — a tenfold increase in speed. The new algorithm could be particularly useful for image compression, enabling, say, smartphones to wirelessly transmit large video files without draining their batteries or consuming their monthly bandwidth allotments.
Like the FFT, the new algorithm works on digital signals. A digital signal is just a series of numbers — discrete samples of an analog signal, such as the sound of a musical instrument. The FFT takes a digital signal containing a certain number of samples and expresses it as the weighted sum of an equivalent number of frequencies.
“Weighted” means that some of those frequencies count more toward the total than others. Indeed, many of the frequencies may have such low weights that they can be safely disregarded. That’s why the Fourier transform is useful for compression. An eight-by-eight block of pixels can be thought of as a 64-sample signal, and thus as the sum of 64 different frequencies. But as the researchers point out in their new paper, empirical studies show that on average, 57 of those frequencies can be discarded with minimal loss of image quality.

Heavyweight division
Signals whose Fourier transforms include a relatively small number of heavily weighted frequencies are called “sparse.” The new algorithm determines the weights of a signal’s most heavily weighted frequencies; the sparser the signal, the greater the speedup the algorithm provides. Indeed, if the signal is sparse enough, the algorithm can simply sample it randomly rather than reading it in its entirety.
In nature, most of the normal signals are sparse,” says Dina Katabi, one of the developers of the new algorithm. Consider, for instance, a recording of a piece of chamber music: The composite signal consists of only a few instruments each playing only one note at a time. A recording, on the other hand, of all possible instruments each playing all possible notes at once wouldn’t be sparse — but neither would it be a signal that anyone cares about.
The new algorithm — which associate professor Katabi and professor Piotr Indyk, both of MIT’s Computer Science and Artificial Intelligence Laboratory (CSAIL), developed together with their students Eric Price and Haitham Hassanieh — relies on two key ideas. The first is to divide a signal into narrower slices of bandwidth, sized so that a slice will generally contain only one frequency with a heavy weight.
In signal processing, the basic tool for isolating particular frequencies is a filter. But filters tend to have blurry boundaries: One range of frequencies will pass through the filter more or less intact; frequencies just outside that range will be somewhat attenuated; frequencies outside that range will be attenuated still more; and so on, until you reach the frequencies that are filtered out almost perfectly.
If it so happens that the one frequency with a heavy weight is at the edge of the filter, however, it could end up so attenuated that it can’t be identified. So the researchers’ first contribution was to find a computationally efficient way to combine filters so that they overlap, ensuring that no frequencies inside the target range will be unduly attenuated, but that the boundaries between slices of spectrum are still fairly sharp.

Read More

Article from MITNews

Tuesday, January 17, 2012

WIAMIS 2012: The 13th International Workshop on Image Analysis for Multimedia Interactive Services 23rd - 25th May 2012, Dublin City University, Ireland

The International Workshop on Image Analysis for Multimedia Interactive Services is one of the main international events for the presentation and discussion of the latest technological advances in interactive multimedia services. The objective of the workshop is to bring together researchers and developers from academia and industry working in the areas of image, video and audio applications, with a special focus on analysis. After a series of successful meetings starting in 1997 in Louvain, WIAMIS 2012 will be held in Dublin City University, Ireland.

All accepted and registered papers will be published in the workshop proceedings which will be indexed and distributed by IEEExplore.

Topics of interest include, but are not limited to:
  • 3D data processing and visualization;
  • Multimedia content analysis and understanding;
  • Content-based browsing, indexing and retrieval of images, video and audio;
  • Content-based copy detection;
  • Emotional based content classification and organization;
  • 2D/3D feature extraction;
  • Advanced descriptors and similarity metrics for multimedia;
  • Segmentation and reconstruction of objects in 2D/3D image sequences;
  • Motion analysis and tracking;
  • Multi-modal analysis for event recognition;
  • Analysis for coding efficiency and increased error resilience;
  • Analysis and tools for content adaptation;
  • Content summarization and personalization strategies;
  • End-to-end quality of service support;
  • Multimedia analysis for new and emerging applications;
  • Multimedia analysis hardware and middleware;
  • Semantic web and social networks;
  • Advanced interfaces for content analysis and relevance feedback;
  • Video/audio-based human behavior analysis;
  • Advanced multimedia applications.

Proposal for Special Session: 9th Dec, 2011

Notification of Special Session Proposal Acceptance:23rd Dec, 2011

Paper Submission: 3rd Feb, 2012

Notification of Acceptance: 30th March, 2012

Camera-ready Papers: 13th April, 2012

Friday, January 13, 2012

3D Photo Ring HD Update v2.3

3D Photo Ring is an iOS app that shows your photos in an interactive three-dimensional arrangement including a color sorting feature. The app also provides an interactive slideshow and can display EXIF meta-data.

Now, a new update has been released, which integrates lots of improvements:
  • Autorotate support
  • Pinch for zooming
  • Color sorting significantly improved
  • Loading of photos more than 2 times faster
  • Aspect ratio problems fixed
  • Switching photos (fullscreen mode) by device tilt
  • Sensibility of rotation speed configurable in settings
  • Persistent storage of settings
  • Automatic rotation of ring on demand
  • Automatic check for changes in photo library on startup
  • Several bugfixes

The iPad version of the app (3D Photo Ring HD) is currently available at half the price.

More information here:

Thursday, January 12, 2012

US Moves Toward Banning Photoshop In Cosmetics Ads

Article from Business Insider

Procter & Gamble has agreed to never again run an ad for its CoverGirl mascara because it used "enhanced post-production" and "photoshopping" to make eyelashes look thicker than they were in real life. P&G agreed to the ban even though it disclosed in the ad that the image was enhanced.

The move is the latest in a series of baby steps that U.S. and international advertising regulators have taken to ban the use of Photoshop in advertising when it is misleading to consumers.

The company's decision was described in a ruling by the National Advertising Division, the U.S. industry watchdog that imposes self-regulation on the advertising business. NAD is part of the Council of Better Business Bureaus. Its rulings are respected and followed by most advertisers because it enjoys a close relationship with the FTC, from which it has historically drawn some of its senior staff. Recalcitrant advertisers who refuse to withdraw or amend misleading ads are referred by the NAD to the FTC, which has the power to fine, sue or bring injunctions against companies.

When asked whether this was a de facto ban on Photoshop, NAD director Andrea Levine told us:

"You can’t use a photograph to demonstrate how a cosmetic will look after it is applied to a woman’s face and then – in the mice type – have a disclosure that says ‘okay, not really.’”

The ad in question was for CoverGirl NatureLuxe Mousse Mascara, which promised “2X more volume” on women's lashes. After reviewing the ad, P&G agreed to yank it. (A different CoverGirl ad is shown here.) The NAD ruling said:

"… [P&G] advised NAD it has permanently discontinued all of the challenged claims and the photograph in its advertisement. NAD was particularly troubled by the photograph of the model – which serves clearly to demonstrate (i.e., let consumers see for themselves) the length and volume they can achieve when they apply the advertised mascara to their eyelashes. This picture is accompanied by a disclosure that the model’s eyelashes had been enhanced post production."

In a footnote, the NAD said it was following the lead of its sister body in the U.K., the Advertising Standards Authority, which in July banned cosmetics ads featuring Julia Roberts and Christy Turlington because they used Photoshop. The NAD said:

"Advertising self-regulatory authorities recognize the need to avoid photoshopping in cosmetics advertisements where there is a clear exaggeration of potential product benefits."

"... the picture of Ms. Roberts had been altered using post production techniques (in addition to professional styling, make-up, photography and the product’s inherent covering and smoothing nature which are to be expected), exaggerating what consumers could expect to achieve through product use."

The U.K. ruling found the use of photo retouching misleading per se.
In the U.S., the FTC has has also tightened rules to hold celebrities accountable if they make claims in ads they know cannot be true.

And in France, in 2009, 50 politicians asked for health warnings to be imposed on fashion ads if they showed retouched models' bodies.

Article from Business Insider


Congratulation to the Klagenfurt team for the organization! Fantastic job.

More Pictures From the Event