Pages: [1]   Go Down

Author Topic: Canon’s Eye-Tracking Experiment  (Read 2640 times)

Slobodan Blagojevic

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 18090
  • When everyone thinks the same, nobody thinks
    • My website
Canon’s Eye-Tracking Experiment
« on: December 10, 2015, 05:57:56 pm »

Quote
Canon has conducted an experiment in which it brought in three people to examine a photograph in order to investigate just what it is that an average person sees in a photograph, particularly as compared to a professional photographer...

http://www.diyphotography.net/canons-experiment-shows-how-obsessed-photographers-are-over-the-details/

walter.sk

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1433
Re: Canon’s Eye-Tracking Experiment
« Reply #1 on: December 11, 2015, 11:09:10 am »

Thanks for the link.  I found it interesting, and having been a composer of music it substantiates my understanding that musical training draws the ear to aspects of the music that the non-musician may also hear but be unaware of.  It also demonstrates the use of image simplicity and impact for advertising, as the non-visually trained magazine reader will spend very little time on the ad photographs.

What I would like to see done, though, is a study that shows the path of the eye as it views the picture, done with several types of image where bright objects, colorful objects, large objects, etc., are placed in different parts of each image.  My experience is that the many of the "rules of composition" that, for example, camera clubs use for judging, and also teach the newcomers, are based more on urban myth than on actual physical studies.
Logged

razrblck

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 482
  • Chill
    • Instagram
Re: Canon’s Eye-Tracking Experiment
« Reply #2 on: December 11, 2015, 11:38:10 am »

Quote
Not only do some of us spend way more time than others making sure every little details is perfect, but these efforts aren’t always even noticed by our audience.

Wow.

Our eyes don't see only narrow points, we have a rather wide field of view. Just because someone didn't bother to check every single pixel of the image doesn't mean those parts are useless or irrelevant to the whole. Besides, using only three people is too limiting to even make any kind of statement out of it. You might get wildly different results with a broader group, but it's whatever because this is just an ad to sell more printers.
Logged
Instagram (updated often)

Isaac

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3123
Re: Canon’s Eye-Tracking Experiment
« Reply #3 on: December 11, 2015, 01:18:42 pm »

a study that shows the path of the eye as it views the picture, done with several types of image where bright objects, colorful objects, large objects, etc., are placed in different parts of each image.

Meanwhile -- "Gaze movements during the beholding of paintings"



My experience is that the many of the "rules of composition" that, for example, camera clubs use for judging, and also teach the newcomers, are based more on urban myth than on actual physical studies.

Quote
"In 1767, … he drew an explicit link between the composition of paintings and the gaze of the beholder. For Diderot, composition is an instruction to the eye, a path which the gaze follows in a certain order.

Eyes do not follow any line of composition in a continuous manner, nor do beholders scan paintings from top to bottom or left to right continously. … the general assumption of art historians … that viewers' eyes move predominantly from left to right - as when reading - when looking at European paintings is empirically incorrect …

However, … several aspects of Diderot's text are indeed correct …"
Logged

walter.sk

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1433
Re: Canon’s Eye-Tracking Experiment
« Reply #4 on: December 14, 2015, 02:27:32 pm »

I checked out the study you linked to, and found it to be thorough in its approach and understanding of what issues are involved in the psychology of perception and cognition when viewing paintings (or photographs).  Strongest was the difference between visually trained subjects (art graduates vs those not interested in art, for example), and then the realization that the visual movements were more determined by the elements of a given painting rather than some more "universal" pattern that had been presumed to exist in the past.

A fantasy I entertain would be going blindfolded into a gallery of a photo club exhibit of, say, 100 pictures.  Feeling the edges of each frame, I would be able to locate the main subject at the lower left intersection of thirds, and report that the right side of each image had leading lines toward the left, with the more solid dark elements at the left edge to "keep the eye from leaving."  Borrrring! 

Thank goodness people are more complex, as well as is the world of art.  While I agree that some of our compositional rules (guidelines) can be helpful, I am afraid that newbies to photoclubs learn to see by those "rules" and then enter a gallery feeling that all of those images were done "wrong." 
Logged

amolitor

  • Guest
Re: Canon’s Eye-Tracking Experiment
« Reply #5 on: December 14, 2015, 07:38:58 pm »

Eye movement is probably not the whole story about how we perceive pictures, but it's an interesting thing to consider, innit?

People look at people, it turns out.

To be honest, I found the advert to be kind of lame. Who cares if the photographer who shot that thing looks all over the frame? Yeah, I probably overthink my own pictures too. The point of the advert ought to be "obsessing over the details is probably silly because ain't nobody gonna look at them, they look at the guy's face and that's pretty much it"
Logged

Justinr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1733
    • Ink+images
Re: Canon’s Eye-Tracking Experiment
« Reply #6 on: December 19, 2015, 05:35:34 pm »

Eye movement is probably not the whole story about how we perceive pictures, but it's an interesting thing to consider, innit?

People look at people, it turns out.

To be honest, I found the advert to be kind of lame. Who cares if the photographer who shot that thing looks all over the frame? Yeah, I probably overthink my own pictures too. The point of the advert ought to be "obsessing over the details is probably silly because ain't nobody gonna look at them, they look at the guy's face and that's pretty much it"

Yep. We all think we are producing our own little masterpieces the deliverance of which is breathlessly waited upon by an adoring multitude. Nothing could be further from the truth!  The devil is in the detail only according to photographers, few others give a hoot which is why its the impact of the image that counts. If the story is not told in a second then forget it from a commercial point of view.
Logged

Zorki5

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 486
    • AOLib
Re: Canon’s Eye-Tracking Experiment
« Reply #7 on: December 19, 2015, 10:49:07 pm »

Uhhh... To put it mildly, this article does not make any sense to me, at all.

The number saccades is dictated by physiology, not any sort of skills. Given equal time to observe an image, people would produce numbers of saccades that differ only due to they neural systems' differences, not anything else.

I would fully expect patterns/tracks of saccades to be different for different groups (such as laymen/amateur/pros). Yeah, I can see that.

Also, patterns of saccades are often used to distinguish well and badly composed images, with the former having very well-defined areas of interest (where the focus stays for much longer), and the latter having the eye jumping all over the place.
Logged

Justinr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1733
    • Ink+images
Re: Canon’s Eye-Tracking Experiment
« Reply #8 on: December 20, 2015, 03:55:13 am »

Uhhh... To put it mildly, this article does not make any sense to me, at all.

The number saccades is dictated by physiology, not any sort of skills. Given equal time to observe an image, people would produce numbers of saccades that differ only due to they neural systems' differences, not anything else.

I would fully expect patterns/tracks of saccades to be different for different groups (such as laymen/amateur/pros). Yeah, I can see that.

Also, patterns of saccades are often used to distinguish well and badly composed images, with the former having very well-defined areas of interest (where the focus stays for much longer), and the latter having the eye jumping all over the place.

And then there is the sample size...
Logged

Phil Indeblanc

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2017
Re: Canon’s Eye-Tracking Experiment
« Reply #9 on: December 21, 2015, 02:31:31 pm »

If Joel Grimes is so obsessed, then why does the image in that promo have a hand look so much larger than the other hand? Is that due to not doing any lens correction? or? Maybe its correct, but it does look odd to me for some reason. If you see it in a magazine print it is much more prominent then in the small video.
« Last Edit: December 21, 2015, 02:34:35 pm by Phil Indeblanc »
Logged
If you buy a camera, you're a photographer...
Pages: [1]   Go Up