Sunday, December 20, 2009

IMAGinE 2010

imagine.jpg

We are happy to invite you to IMAGinE (read: Imagine in IMAGE),
an informal, non-commercial gathering of like-minded people in the fields of animation, computer vision, 3D graphics, image processing, visualizations, augmented reality, video and photography.

IMAGinE is one of the unconference events initiated and inspired by Yossi Vardi, which include Kinnernet, GeekCon and MuseNet. As an unconference, the content is created by the event participants, so we depend on YOU to propose discussions, workshops, project displays and any crazy mash-up that comes to mind and can drive technologists and creative types to innovate, learn and inspire.

IMAGinE 2010 will be held on Friday, January 8th 2010, at the Tiltan College of Design in Haifa.

The event wiki is located at http://imagine2010.pbworks.com, and registration is at http://imagine2010.eventbrite.com.

facebook event at http://www.facebook.com/event.php?eid=210873607370

See you there!

The IMAGinE organizing team: Dror Gill, Liad Nishlis, Erez Yissacharov and Yossi Vardi.

Wednesday, November 04, 2009

Yet Another Tron Custome


Yes, there are few more geeks who made a Tron costume,
but none of them built a frisbee controller and played the arcade game Discs Of Tron.

More info and pictures at the Tron costume project page

Wednesday, October 28, 2009

MakeUseOf.com review




Have you ever wondered what your subscribers are seeing when they come to your website or blog? Feng-GUI predicts what a human mind would look at within 5 seconds of looking at a web page and generates visual heatmap. This is the perfect way to tell what your readers are watching and what is grabbing their attention.

read the full review at

http://www.makeuseof.com/dir/feng-gui-visual-heatmap/




MakeUseOf also listed Feng-GUI at cool websites and tools

Wednesday, October 21, 2009

Moving to Chicago on November 1st


At 1st of November we are migrating the Feng-GUI Web Service to Chicago.
Our neighbors at the Data Center are PHP, YouSendIt and Ars Technica.

Dear Customer,

Please be informed that essential maintenance has been scheduled on the Feng-GUI services infrastructure on:
Sunday 1st November between 8.00 and 10.00 GMT+2.

During this time our Engineers will work to minimize any disruption to the Feng-GUI services, nevertheless please be aware that the planned maintenance is taking place.

Best regards,
Feng-GUI Support

Wednesday, September 09, 2009

Service Level Agreement published


The Feng-GUI Service Level Agreement published.
The agreement guarantees 99.9 percent availability of the service within a region over a trailing 30-day period. If Feng-GUI Services fails during that time, or doesn't meet the agreement, businesses get credit back.
"... the Feng-GUI Covered Services web interface will be operational and available to Customer at least 99.9% of the time in any calendar month (the "Feng-GUI SLA"). If Feng-GUI does not meet the Feng-GUI SLA, and if Customer meets its obligations under this Feng-GUI SLA, Customer will be eligible to receive the Service Credits described below."
"If the Monthly Uptime Percentage applicable to the month of such request is confirmed by us and is less than 99.9%, then we will issue the Service Credit to you within one billing cycle following the month in which the error occurred."


here is an example of an email sent from Feng-GUI support to a customer, with a credit back notification.

Sunday, August 09, 2009

Whats new in Dashboard V2?

14th July 2009, Feng-GUI Dashboard V2 released.
here's some information on what's new in this release.

Loading time is 5 times faster.

The most recent generated images are presented in small thumbnails. These small images makes the page load 5 times faster.



Face and Skin detection
Increasing the accuracy of the attention heatmap by adding frontal and profile face detection along with skin hue detection.
Eye-tracking researches has proven that during the first half of a second we are attracted to faces found within the image.

Customized output layers
Choose which information to include at the output attention heatmap.



Transparent credit system
You can see how many credit you have at the Account Settings tab.



You only pay per image upload
We separated the upload and the heatmap into two actions.
Previously, if you wanted to test the same input image using different analysis parameters, you had to upload the same image every time you generate a heatmap.
Now, you are required to upload the input image only once, and for that image, you can generate as much heatmaps as you want.



And there’s another good thing for the Dashboard customers. You are charged only for the upload action, and
not for the heatmap action.
Notice, This does not apply to Web API users.
So this is how it goes. Click on ”Browse” button, to select an image file from your computer. Click the “Upload” button to upload it to the server.
Now, select the desired image from the thumbnails filmstrip at the bottom of the page, and click “Heatmap”.
This way, you can change the Analysis parameters and generate as much heatmaps without upload the image each time.

Saturday, July 11, 2009

Next Generation Digital Media Survival Kit

Following the Innovation Summit at Berlin, Wolf Siegert Ph.D. from IRIS Media (http://www.iris-media.com/) came to Israel looking for the Next Generation Digital Media Survival Kit. Wolf was interviewing for 3 days various New Media companies including Feng-GUI




Here is part of his report auto translated from German to English

This Sunday is - in Israel - a working day.
The theme - Next Generation Digital Media Survival Kit - this was three days on the Israel Export & International Cooperation Institute in Tel Aviv as follows in his "follower" by electronic bulletin announced:
One of the founders of this GarageGeeks - concept was already the day before the host of the Export Institute - now he is host.



His name Rafael Mizrahi shortened in speech-friendly on "Rafi, the job of the author is on the" changineers "and it will be shortened by" Rafi "with a very special welcome honors:
The particular pleasure of this meeting is in the fact that AOL's launch at the IFA in Berlin, the mid-90s in collaboration with the invitations took place today, and today - even the AOL brand - to the Telecom Italia was sold.
In advance of the event allowed Rafi, at the end of the room and settle the computer presentation.

Evening, half of new, suddenly met a big crowd of people and could be increased before this small work table.
And then began a lot of laughter during miscellaneous presentation of all sorts of nonsense, its deeper meaning is often not until the opening statements.
Around the bicycle with a motor, which does nothing more than to make noise and the environment to "contaminate".

Or snail mail to deal with CD-Rom wheels move attempt (even a "Snail mail" is nimble on the other hand ...)




After the exclusive presentation in the backyard Grage, participants gathered in the backyard and listened to Tim Armstrong keynotes.

Wednesday, May 27, 2009

Master Geeks Hagai and Mel

 

Hagai Cohen and prof. Mel Rosenberg are true master geeks.
Check out their two projects A compressor flute organ and a land skiing vehicle.
These guys are old school hard core! A true inspiration for us; the next generation.

Sunday, May 17, 2009

Put your ear on the View Finder

Alexey Klimov (Алексей Климов) , photography expert and researcher, is using Feng-GUI to identify aesthetics in Synaesthesia photos, taken “from his ear” !

Feng-GUI have kindly given me possibility to analyze my exhibition photos.
I have use Feng-GUI to automatically search for pictures who most pleased the viewers. Almost all from them have shown a similar arrangement of zones of attention (hetamap). Thus, that the photographer style can be calculated without the aid of man-critics.


Synaesthesia - for some people letters are colored, numbers have a personality and when they hear music they see colourful images.
Synaesthesia explained at BBC http://news.bbc.co.uk/2/hi/7545888.stm









"I have started to photograph when I was eight-year old. It was the time of a black-and-white film and a house “red room” photo laboratory.
Since then I make at least one shot daily. I have read many books of known photo masters. I studied at art school.
But I have chosen other trade. The photo became my hobbies. Surprisingly, but my academic classical photos bored the viewers.

All has changed, when my wife has presented to me a mobile phone with the built in camera.
I began to photograph "from an ear" during telephone conversations. I aimed on hearing. Sight and all academic knowledge had a rest during such photographing. The man gets used to all. In three years I consider it natural to put the view-finder not my an eye, but to my ear. From hundred shots one turns out adequately. Such perfect shots it was typed on the whole photo-exhibition.”







Alexey Klimov Synaesthesia exhibition photos
http://www.foto3x4.narod.ru/Sinestes.html
http://foto3x4.narod.ru/Muzey.html

Today I have developed ability "to photograph from an ear" to level of regular " view-finder photographing". Viewers began to speak about photographer's style. Experts began to speak about mixture of feelings (Synesthesia). The doctor of neurosurgery marks effect of dreams: when our eyes are closed we have dreams, just like babies. After all, when we dream, hearing is not blocked as sight.

Unexpectedly, I have found out Feng-GUI which helps me to automatically analyze my photos. Using Feng-GUI I have found a way to select the photos which most please the viewers. Now my Feng-GUI tested photos are printed in several Russian magazines. Photos which I take "from the ear" began to be attractive.



We like to see ourselves in group photo. If we don't find out ourselves in a photo then we value it as a documentary or art one.
When we see people in photo, the first thing that comes to our mind is 'Do I know them?' and the second is 'Who are they?'
The city life has accustomed us to see thousand strangers every day.
Our brain doesn’t store flashing persons in long–term memory to protect itself against overload.
And this habit is transferred also to taking photo.
Moreover, we distinguish people in our dreams on familiar or strangers, though we can’t remember their features and details after waking up.
For one thing, could you describe their ears, nose, eyes and lips if it was required to make an identikit for criminal investigation? :)



And still most of us, while having got a city intelligent habit 'don’t examine', 'don’t stare at stranger' and so on, remained wild animals in a our souls, and we expect everything of each stranger.
That’s why we learned to monitor life space about us by off–center vision, and using hearing and tactile perception in transport.
So how we can portray this point of view in photo?
Right in this manner, off–center vision–like.
You have to construct a shot so that the spectator didn’t look at an unfamiliar portrait, and was mown, intuitively listened, or even sniffed it.
So he didn’t stare at the person, like he’s looking at a monkey behind glass.
And the mathematics, 'queen of all sciences', allows to make it.

The first impression of a man is the strongest.
It is a wonder why nature and society together have developed this special first glance for creating a first impression of a stranger.
We do not stare straight at a person at the first encounter (unless we work in an HR department).
However, quick glances at the face, peripheral, panoramic view of a stranger's position in the interior, the elevated attention to information from all our senses — this is what provides the very first impression.
How to convey the first impression with a photo?
Especially, if the model is in fact well familiar to you?
The viewer does not have the intuitive barriers against staring at the monitor or the pages of a photo album.
The viewer can shamelessly eye a stranger's photo, feeling secure.

How to convey to the viewer the anxiety that he would experience if he met in person the man in the photo?

About the author
Alexey Klimov, a resident of the city of Oryol, is the author of over fifty publications on digital photography in magazines “Computerra”, "Home Computer" and “Hard and Soft”. Candidate of technical sciences, associate professor, Ph.D at Oryol State Technical University
http://www.computerra.ru/Authors/793/offline.html
http://www.computerra.ru/magazine/357229/

Thursday, April 30, 2009

JSON joins the services supported formats

This one is for you AJAX lovers out there.
Feng-GUI web services now supports JSON (JavaScript Object Notation).

You may find the JSON end point at http://service.feng-gui.com/json/api.ashx

Note: Only registered customers can access the web services.

Thursday, February 26, 2009

Lost in Conversion?

LostOnConversion

Lost in Conversion? Unleash the power of web analytics to improve conversion rates.

Web analytics and conversion optimization is probably the biggest untapped opportunity within online marketing, particularly in a time of economic downturn.

Within this GarageGeeks session, you be exposed to best practices and new developments in measurability, user behavior analysis and proven conversion rate tactics.

Conversion is the New Black!
Itai Levitan, Co-CEO and Daniel Waisberg, Head of Web Analytics, easynet search marketing http://www.easynet.co.il

Google Analytics & Website Optimizer an Overview
Nir Bar-Lev, Head of Analysis Products, EMEA at Google

From Porn to AI in 20 slides.
Rafael Mizrahi, http://www.Feng-GUI.com

Reveal the mystery of what visitors are doing inside your webpage
Dr. Tal Schwartz, Founder and CEO, ClickTale
http://www.clicktale.com

NuConomy - next generation personalization and advertising platform based on advanced user level Analytics.
http://www.NuConomy.com

Session Summary
Itai Levitan of easynet search marketing.

- Mingle and Mangal

The event will take place on Thursday, March 12th, 20:00, at the GarageGeeks HQ (Hapeled 40, Holon).

Thursday, February 05, 2009

Digital Innovators Summit



Feng-GUI is invited by the 'New Media' Sector at the Israel Export Institute to present at the Digital Innovators Summit, Berlin.
We will be presenting at the "Innovative Companies: Analytics and Ad Management" Panel, with Microsoft Advertising and ClickTale.


VDZ, the Association of German Magazine Publishers, along with
www.emediasf.com and CeBIT are hosting the "Digital Innovators Summit" which bring together dynamic companies to present their digital media concepts and offerings.
The conference will be held on March 3-4, 2009 at the Deutsche Telekom Conference Center, Franzo"sische Strase 33a-c, 10117 Berlin, Germany.

Feng-GUI welcome the "Digital Innovators Summit" participants.
You can access the Dashboard using the Feng-GUI XNIP-Code (as username/password) found at the summit program.

Tuesday, January 20, 2009

Real Virtual Skater


Another release is behind us, so this is time for some geeky stuff.
We took a real skateboard and converted it into a game controller .
Taking a real skateboard + a Wii remote and using it to control and "ride" the game "Tony Hawk Pro Skater" (THPS)

http://www.feng-gui.com/thps/

more projects at the projects page
http://www.feng-gui.com/research.htm

Thursday, January 08, 2009

Slip me some skin, Bro!

Feng-GUI free attention heatmap service now includes skin detection.
Skin and Face detection have an important role in human visual attention.
It was already demonstrated in eye tracking sessions that face and skin are the first and the second spots that people tends to look at.

Checking “skin detection” will employ a combination of skin models and statistics to identify skin hue color in your image.

Do not use it unless you know that there is skin in the image.


original image

attention heatmap with no skin detection

attention heatmap with skin detection

Monday, January 05, 2009

Feng-GUI attention heatmap accuracy

We’ve been asked several times about the accuracy of Feng-GUI attention heatmap and why is it not reaching 100%.
Feng-GUI attention heatmap algorithm, which we call the ViewFinder , is a composition of algorithms aggregating over 30 years of research. Research from the fields of neuro-science studies, visual attention, statistics of eye-tracking sessions, perception and cognition of humans.
We improve the ViewFinder algorithm in order to achieve the highest accuracy, but is it possible to get to 100% ?

Let us answer by asking a question:
How accurate is the "Magic Wand" in Photoshop? (also referred as the “Fuzzy select” in Gimp)



The Magic Wand tool is designed to select areas of the image by finding edges of image segments.



Part of the answer to accuracy can be found at the "image segmentation and boundary detection" benchmark conducted by Berkley.
It’s a 1000 photos test suite for algorithms that try to find the boundaries of elements inside pictures.
In the test suite results you can see how real people and computer algorithms, marked the boundaries of segments inside photos and how accurate they are.
The current score is:
Real people: 79%
Machines algorithms: 70%


original image

edge detection algorithm

edge detection by human

http://www.eecs.berkeley.edu/Research/Projects/CS/vision/grouping/segbench/
results:
http://www.eecs.berkeley.edu/Research/Projects/CS/vision/grouping/segbench/bench/html/algorithms.html

You can see that edge detection algorithms are not 100% accurate and will never be considering the hardware used be personal computers.
To solve this limitation, GIMP supports six different selection tools. In order of appearance in the tool box they are: rectangle, ellipse, free, fuzzy, bezier, and intelligent.
Each tool has its advantages and disadvantages--no one tool is clearly superior to the others. In general, a combination of tools proves to be the optimal strategy for complex selection tasks.

We conducted a comparison benchmark between Feng-GUI results and eye-tracking results of 40 people, 90 images, 15 seconds per image.
http://www.feng-gui.com/faq.htm#quality
Comparing the regions of interests (ROI) in the images, results with more than 70% of similarities.
If you compare Feng-GUI results to random groups of 10 people (out of the 40 people), for 5 seconds, the similarities go over 80%.

To enable designers and advertisers integrate Feng-GUI’s attention heatmap into their work process, we keep on improving the ViewFinder algorithm while maintaining two important requirements: Accuracy as well as Speed.

happy new year!