Clever Geek Handbook

Why is Google Face Recognition only available on Picasa and not on Picasa for PC?

I have a friend asked me today.

Picasa Web has a cool (and scary :-) feature where it recognizes all the faces in your photo album.

But the PC version (desktop) doesn't have that.

Several reasons I can think of:

  • They just didn't have time to write the PC version of the code.
  • They license this feature and it costs a lot more (or not available) on a PC.
  • Takes a lot of processing power (it seems odd b / c MY PC cycles are free for Google, but they have to pay for the cycles consumed on their server.

Any other thoughts?

+1


source to share


6 answers


I'm sure this will happen in future releases, but Google is a fun company when it comes to its own competing / complementary services. One thing is for sure, only someone from the Picasa team could have given the exact answer.

But we could assume a few things ...



  • They don't want their code to be reverse engineered.
  • (As you say) they don't have a redial license.
  • It is locked in the dev version by other new features not yet completed.
  • They don't want to release it because they want people to use PicasaWeb as a social photography network.

I don't think computing power is an issue. If they download it for free on their servers for free, a modern desktop might run it without issue.

+4


source


From my limited contact with facial recognition software, this is probably a redistribution issue. When I dealt with this, facial recognition was its own little world with extremely high processor licensing costs and huge paranoia about the code being made free.



+2


source


I'm not sure if this is not a processing issue. It took 30 million Google servers 30 minutes to view all my photos. I can only imagine that the same task would take days on my local machine.

+2


source


In fact, inside it, only in limited functionality, when you do a search, there is an icon to find only photos with faces. The experimental passport function also works this way.

So the answer is:

  • Not the same base (API) available or used, not the same language, so it is not directly portable.
  • Not the same software and there are no stated goals to make both applications equivalent.
  • Programmers are limited, and so is their time. They make choices about what they are now implementing.
+1


source


I don't know if this is the case for Picasa, but there is another case where a licensing issue can be a problem. If the server-side code uses code with a restrictive DRM license (such as the GPL), which restricts the distribution of modules with the code. Running this module on a web server where the user only gets the output is legal under such licenses. If this code were distributed, there would be many legal requirements that would probably be highly undesirable for commercial software companies, including google. This is one very good reason to have some of the capabilities available only through web services.

0


source


It was also with Riya (who may have been the first to market with robust facial recognition for consumer photo collections).

The biggest reasons are possible:

  • Processing time (they cannot control how fast your processor is and therefore they cannot control the experience). Facial recognition is very likely to be an intensive process (this was Riya's stated reason not to do it on the client side)
  • The recognition process requires a LARGE amount of data for processing which is available only on the server? (In other words, the process needs to loop through millions of faces, not just the faces you have on your hard drive?)
0


source






More articles:


All Articles