Comments on: Feng-GUI Heatmap: what are people look at? http://www.markpascua.com/2008/03/26/feng-gui-heatmap-what-are-people-look-at/ Look at what I found! Tue, 16 Mar 2010 07:44:47 -0700 http://wordpress.org/?v=2.9 hourly 1 By: Rafael Mizrahi http://www.markpascua.com/2008/03/26/feng-gui-heatmap-what-are-people-look-at/comment-page-1/#comment-712 Rafael Mizrahi Wed, 26 Mar 2008 13:41:43 +0000 http://www.markpascua.com/2008/03/26/feng-gui-heatmap-what-are-people-look-at/#comment-712 Hi Mark, thanks for sharing Feng-GUI with your readers. regards Feng-GUI accuracy. we improve the ViewFinder algorithm in order to achieve the highest accuracy, but is it impossible to get to 100% ? Let me answer by asking a question: how accurate is the "Magic Wand" in Photoshop? Part of the answer can be found at the "image segmentation and boundary detection" benchmark conducted by Berkley. Its 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: 67% 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 We recently published a comparison report 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 The results were 70% of matching hotspots regions. If you compare Feng-GUI results to random groups of 10 people (out of the 40 people), the match to 80%. Feng-GUI will keep on improving the ViewFinder algorithm while maintaining two important requirements: accuracy and speed. stay tune. and thank you very much. Hi Mark,
thanks for sharing Feng-GUI with your readers.
regards Feng-GUI accuracy.
we improve the ViewFinder algorithm in order to achieve the highest accuracy, but is it impossible to get to 100% ?
Let me answer by asking a question: how accurate is the “Magic Wand” in Photoshop?
Part of the answer can be found at the “image segmentation and boundary detection” benchmark conducted by Berkley. Its 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: 67%

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

We recently published a comparison report 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
The results were 70% of matching hotspots regions.
If you compare Feng-GUI results to random groups of 10 people (out of the 40 people), the match to 80%.

Feng-GUI will keep on improving the ViewFinder algorithm while maintaining two important requirements: accuracy and speed.

stay tune. and thank you very much.

]]>