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

Labels: ,

posted at 11:23 AM

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!

Labels: , ,

posted at 1:24 AM

Tuesday, December 16, 2008

November reviews 2008

posted at 11:18 PM

Tuesday, November 25, 2008

SOAP joins the supported formats

We are glad to announce that SOAP format joins the supported formats of Feng-GUI web services.

http://www.feng-gui.com/services/
http://www.feng-gui.com/services/soap.htm

coming up next, JSON (JavaScript Object Notation) ...

Labels: ,

posted at 10:42 AM

Tuesday, November 04, 2008

is your website ready for the 5 seconds attention span challenge



After decades of TV, Radio, Magazines and now Internet, we find ourselves in a society overwhelmed by information begging for attention. As a result of that we have a world with an attention span now shorter than a goldfish’s. How is the media and specially the web facing that? And how can we take advantage on way the human mind works?

read the wonderful post by Mauricio Duque at snap2objects

Thanks!Mao

Labels: , ,

posted at 1:17 PM

Wednesday, September 24, 2008

Optimizing AdSense using Feng-GUI heatmap


Alex Barbadosa, frmo http://www.lessnau.com the AdSense blog, twits that
I changed the position of www.riojob.com.br Adsense on my CRT and jumped from 0.5% to 3.50%. Establish in-www.feng gui.com (Heat Map) evolving ...


optimize your AdSense location and style using Feng-GUI heatmap.

Labels:

posted at 10:49 PM

Sunday, June 15, 2008

Percentage of Web Page Text Read

useit.com

on How Little Do Users Read?
by Jakob Nielsen's Alertbox, May 6, 2008:
On the average Web page, users have time to read at most 28% of the words during an average visit; 20% is more likely.


read the whole article at http://www.useit.com/alertbox/percent-text-read.html

Labels: ,

posted at 3:41 AM

Sunday, June 01, 2008

Google patents media layout analysis

Google just published two patent applications. US2008/0107337 is “Methods and Systems for Analyzing Data in Media Material Having Layout” and US2008/0107338 is “Media Material Analysis of Continuing Article Portions”. You can view them at USPTO.

Both inventions, to which Google is the assignee, pertain to figuring out what’s important and what’s not on Web pages. Companies that scan hard copy and convert those images to machine-readable ASCII use some tricks but a great deal of brute force to figure out what’s information and what’s advertising or other dross.


thanks to Stephen Arnold for the information.

Labels: ,

posted at 6:11 AM

Brain Scanner Can Tell What You're Looking At

Scientists have developed a computer model that predicts the brain patterns elicited by looking at different images -- a possible first step on the path to mind reading.


Image: University of California at Berkeley
read more at http://www.wired.com/science/discoveries/news/2008/03/mri_vision

Labels: ,

posted at 2:03 AM

Thursday, May 15, 2008

DisrupTV convention



Feng-GUI for Videos will be demonstrate at DisrupTV convention (23 May). We will show how to predict people's attention in videos, commercials, and games.

Disrupt.TV is an un-conference intents to bring together Israeli and global Technology, content and creative talents that are taking part in creating the future of TV and media for sharing knowledge, exploring the opportunities and to spark new ideas about how TV and Video will change us and we will change TV.
It is aimed for people who are in the conjunction of Media, Content, Creative and Technology.

Labels: ,

posted at 1:17 AM

Tuesday, May 06, 2008

Contextual vs Behavior Targeting


An Eye Tracking study designed and analyzed by Next Century Media was carried out by PreTesting and TACODA.
That study shows that Behavioral Targeting averages 17% more Ad Looks than Contextual Targeting, and this average increases to 54% more Ad Looks after the first exposure.
In other words, as frequency builds, the same ad increasingly gets filtered out in the Contextual environment, but apparently as a result of surprise, the level of Ad Looking stays up in Behavioral.




The first Eye Tracking study ever conducted comparing Behavioral Targeting to Contextual Targeting finds that, on average, the same ads receive +17% more Looks when seen in unrelated-content sites, than when seen in sites specializing in the sale of those types of products.

After the first exposure, this +17% advantage for Behavioral over Contextual Targeting increases dramatically to +54%.

The fact that Behavioral Targeting, unlike Contextual Targeting, tends to build with frequency, suggests that there is a Surprise effect and that it does not wear out, and because of this repeat Surprise effect, the ad is given more chances to engage the user.


you can view the full report at
http://www.tacoda.com/success/research.php

TACODA conducted primary research to identify the demographics and behaviors of online 'heavy clickers.'
Who are the "Natural Born Clickers" and what is "the Invisible Hand" ?




The invisible hand is a metaphor coined by the economist Adam Smith to illustrate how those who follow their individual self-interests inadvertently stimulate the economy and assist society as a whole. There has been an invisible hand (or set of hands) that through self-interest have guided the development of online advertising, stimulated it through their innate curiosity and sometimes impulsive natures. This is the tale of a group of people who have had a profound impact on our lives as media planners and buyers. We've never known who they are or where they come from. We just knew that they were there, invisible, hiding within the numbers that make up our campaign report


you can view the video presentation and download the report at
http://www.tacoda.com/success/research.php

Labels: ,

posted at 4:17 AM

Tuesday, March 04, 2008

35-50 people for valid eye tracking


Mihkel Jäätma, an Objective Digital partner from Realeyes.it, has created this compelling example that shows you need 35-50 people to draw valid conclusions from eye tracking. However, he holds that useful information can be gained from fewer participants.

Thanks to James Breeze for the information.

Labels:

posted at 1:15 AM

Tuesday, February 12, 2008

Web Vision explained by StomperNet


StomperNet released a bunch of great videos about Web Vision describing how do we See the web while surfing.
"What web designers should know about human vision! Not just eye tracking, but eye gaze simulation and an interactive experiment to help you get it."
for example, take a look at http://youtube.com/watch?v=-SCtjwK8PAY


watch all the videos at http://youtube.com/user/stompernet

They also provide a free (Adobe AIR based) tool called Scrutinizer that offers a simulation of the human visual system. Specifically, it illustrates the distinction between foveal and peripheral vision in visual acuity and color perception.

Labels: ,

posted at 7:18 AM

Sunday, January 13, 2008

Comparing Eye and Mouse tracking with Computational Attention


Matei Mancas, PhD focus his research at the Faculty of Engineering, Mons, Belgium (FPMs) on Computational Attention.
His research demonstrates several interesting applications that employ computational attention.
The proposed applications are from various areas such as signal processing, data optimization etc'
Matei Mancas Computational Attention page

Matei also developed with two of his students Simon Wallerand and Fre'de'ric Lavis, a mouse-tracking technique he used as a a validation tool for his thesis.
You can participate online in his research by adding your own mouse (that follows your eyes) movements to the research results.
You can upload your images and get the mouse-tracking results on these images. You may also upload entire sets of specific images and than ask for the top-down model to the website administrator.

Labels:

posted at 11:03 AM

Thursday, January 10, 2008

Exploring human eye behaviour using a model of visual attention

Oyewole Oyekoya and Fred Stentiford presented at the 17th International Conference on Pattern Recognition (ICPR) 2004:

It is natural in a visual search to look at any object that is similar to the target so that it can be recognized and a decision made to end the search. Eye tracking technology offers an intimate and immediate way of interpreting users' behaviors to guide a computer search through large image databases.

This work describes experiments carried out to explore the relationship between gaze behavior and a visual attention model that identifies regions of interest in image data.
Results show that there is a difference in behavior on images that do and do not contain a clear region of interest.

the rest of the article at ieee

Labels:

posted at 1:07 AM

Monday, December 17, 2007

Artists look different

When doing Eye-Tracking survey, don't include artists, they may change the results by 20%.



A Norwegian study, that showed 16 pictures to both trained and untrained artists used eye-tracking software to show that not only do they see the world differently when drawing it, they also see differently when studying it.”



Read the full story at "Cognitive Daily":
http://scienceblogs.com/cognitivedaily/2007/03/artists_look_different.php

Labels:

posted at 6:34 AM

Monday, October 22, 2007

Internet users quick to judge

By Judy Skatssoon for Science Online

Internet users can take just one-twentieth of a second to decide whether they like the look of a website, researchers say.

Dr Gitte Lindgaard and colleagues from Carleton University in Ottawa flashed up websites for 50 milliseconds and asked participants to rate them for visual appeal.



When they repeated the exercise after a longer viewing period, the participants' ratings were consistent.

"Visual appeal can be assessed within 50 milliseconds, suggesting that web designers have about 50 milliseconds to make a good impression,"

the Canadians report in the journal Behaviour & Information Technology.

Associate Professor of psychology Bill von Hippel, from the University of New South Wales, says it takes about 50 milliseconds to read one word, making this a "stunningly remarkable" timeframe in which to process the complex stimuli on a website.

"It's quite remarkable that people do it that fast and that it holds up in their later judgement," he said.

"This may be because we have an affective or emotional system that [works] independently of our cognitive system."

He says that in evolutionary terms, this ability helped us respond rapidly to dangerous situations.

full article at ABC News

Labels:

posted at 4:14 PM

Tuesday, August 28, 2007

Banner Blindness by Dr. Jakob Nielsen


An article by Dr. Jakob Nielsen from http://www.useit.com

The most prominent result from the new eyetracking studies is not actually new. We simply confirmed for the umpteenth time that banner blindness is real. Users almost never look at anything that looks like an advertisement, whether or not it's actually an ad.


Users rarely look at display advertisements on websites. Of the four design elements that do attract a few ad fixations, one is unethical and reduces the value of advertising networks.


read the full article...

Labels: ,

posted at 1:14 AM

Tuesday, May 15, 2007

Good-Gaze Attention heatmap


Stan, from lijit, just mailed me about a new Visual Attention service Germany called Good-Gaze.

goodgaze team are from the Cognitive Science department at Osnabrück university, Germany.

I have registered good-gaze service, and looking forward to see their heatmaps.
thanks! Stan

Labels:

posted at 2:43 PM

Wednesday, May 02, 2007

Surprising Studies of Visual Awareness

viscog - VisCog Productions and Visual Cognition Lab have released this amazing dvd which includes the famous "gorilla/basketball" video in 2003.

More clips can be found at the Lab:
http://viscog.beckman.uiuc.edu/djs_lab/demos.html

Labels:

posted at 4:26 AM

Wednesday, April 25, 2007

We need a Digital Cortex


Stan James (founder and CTO of Lijit Networks)
on how the world of information need a digital cortex, and the Attention role in the relations between consumer, publisher and advertiser.

Attention to dollars, and other exchanges

We need a Digital Cortex

Digital Cortex 2 - Information overload in the brain

Labels:

posted at 12:11 AM

Tuesday, December 26, 2006

Open Source clicks heatmap apache module


David from http://www.corunet.com/ published the source code of an apache module which analysis the web server logs and produces heatmap images of the web pages.
enjoy and deploy.

Labels:

posted at 1:42 PM

Sunday, December 24, 2006

Eye Tracking and clicks services

Several consulting companies currently sell evaluation services using eye tracking to graphic design houses and web content creators among others.

http://www.clicktale.com

http://www.eyetools.com

http://www.factone.com

http://www.veridicalresearch.com

http://www.tobii.com/

http://www.simpleusability.com

http://crazyegg.com/

http://www.interactive-minds.com/

http://www.clickdensity.com

By ANTHONY SANTELLA:
THE ART OF SEEING: VISUAL PERCEPTION IN DESIGN
AND EVALUATION OF NON-PHOTOREALISTIC RENDERING

Much of the knowledge above about human eye motion has been gained through the use of eye-tracking. A system measures a viewer’s eye in one of several manners and records the point where it is looking, termed the point of regard or POR.

Eye position has been used as a cursor for selection tasks in a GUI [Sibert and Jacob, 2000]. They have also been used to indicate a users’ attention to others in a videoconferencing environment [Vertegaal, 1999].
Another class of use, related to ours, uses POR to control simplifying images or scenes for efficiency purposes. Knowing where a user looks enables pruning of information that is not perceptible, and need not be transmitted in a video stream [Duchowski, 2000].

Eye tracking has been used to evaluate the effectiveness of informational
displays including application interfaces [Crowe and Narayanan, 2000],
web pages [Goldberg et al., 2002], and air traffic control systems [Mulligan, 2002]. As mentioned earlier, eye movements may even reveal information that viewers are trying to report, but cannot, because it is not consciously available.


Experiments have shown that professional radiologists examining slides look longer at locations where tumors are present, even when they fail to identify and report them [Mello-Thoms et al., 2002].
In the future, this might hold the promise of computer assisted technologies to avoid such mistakes.

Labels:

posted at 2:20 PM

Monday, December 11, 2006

The Usability Professionals Association

UPA Israel is a community of Israeli professionals in the fields of usability, user interface design, human factors engineering, user experience design and the gray areas that lie in between.
http://upaisrael.org/English.htm
member of the The Usability Professionals Association

Labels:

posted at 3:41 AM

Tuesday, October 17, 2006

T-Blog call for alpha testing program

T-Blog, the T-Shirts designers blog, explains (in hebrew) about Feng-GUI and calls for designer to join an alpha testing program.
more info at:
http://www.printmall.co.il/blog/Default.asp?Msg_ID=90

Labels:

posted at 6:47 AM

Thursday, October 12, 2006

CrazyEgg website tracking service

CrazyEgg website tracking service, track, evaluate and improve your website.
Get a clear picture of where your visitors are clicking and enhance your site's results. With AJAX like click overlays, lists and heatmaps showing what on your site is popular and might need improvement. Crazy Egg is designed to help you continually test and improve your site.

I'm not sure I understood their service.
Is it an alternative or an enhancement for analyzing the web server logs ?
Doesn't Google Analytics sufficient?

check it out:
http://www.crazyegg.com

Labels:

posted at 2:01 AM

Saturday, August 26, 2006

Designing Usable & Visually Appealing Web Sites

Great list of rules of thumb for Designing Usable & Visually Appealing Web Sites.
created by Wayne Neale and Cindy McCombe from Kodak.
http://acm.org/sigchi/chi97/proceedings/tutorial/wn.htm

Labels: ,

posted at 11:57 AM

Tuesday, August 22, 2006

DreamWorks Shrek2 rendering QA


Hector Yee, who also works for DreamWorks animation, has published at SourceForge, the image comparison component he used for testing Shrek 2 renderer engine.
Perceptual Image Diff (pdiff) - an image comparison utility that compares two images using a perceptual metric.
That is, it uses a computational model of the human visual system to determine if two images are visually different, so minor changes in pixels are ignored.

http://sourceforge.net/projects/pdiff

Labels:

posted at 3:49 AM