Last week's Picasa software update from Google brought with it a neat trick--facial recognition. But it wasn't the first free consumer photo-editing software to find faces. In January, Apple unveiled the latest version of iLife, which included an updated version of iPhoto that could detect and recognize faces in your photos. And this time last year, Microsoft released an updated version of its Windows Live Photo Gallery desktop software that could find faces inside of photos, though it couldn't (and still can't) recognize who's in them.
So, how do these three stack up? To figure that out, we put them to the test. Using 500 sample photos on fresh installs of each program, we tracked around how long each of the tools took to process all the photos, as well as some notable hits and misses from each.
To be fair, our results may not scale, or match the experience you will have. For one, we're using a test bed of photos that's almost entirely 12-megapixel JPEG files, whereas some people may be shooting smaller or larger files that may be in different formats and contain large groups of people--something that can slow these programs down. You're also likely to have a whole lot more than 500 photos sitting around on your computer; we certainly do.
Note: Adobe's PhotoShop Elements software (for Windows | Mac), which also includes a facial recognition feature was not included in this roundup since it's a paid application. Technically iPhoto is as well, but we included it since it comes free on all Macs.
The apps and workflows
iPhoto
iPhoto is the only product of the bunch that's Mac-only. It comes bundled with all new Macs, but the latest version (which includes face detection) must be purchased as a software upgrade if you've got iPhoto '08 or lower. We've included it in this roundup as a free product since it comes bundled with all new Macs.
Face scanning in iPhoto happens automatically, but it's largely a manual process, requiring users to "train" the system to recognize certain faces. The program took around nine minutes to scan through our 500 test photos and when it was done it didn't offer up any suggestions of photos with faces in them.
Instead, users are required to click on a photo with a face in it and hope the program picked it up. If it has, users can simply type the name in--which will auto complete if the person is in your Mac address book. If someone's face was not found, but you can see it in the photo, you can manually contain the face inside of a box, then tag it with their name.
iPhoto's system offers up suggestions of faces it thinks belong to certain people.
(Credit: CNET)After you add names to just few photos, iPhoto's system begins to piece together others that look the same--although it doesn't learn as fast as it does for photos where it already found the faces. In my testing, it only took two photos to get it to offer up some more suggestions. If those suggestions are correct, continuing to add them was just a matter of a few clicks.
iPhoto's system for doing this isn't perfect though.... Read More
Roughly a year after rolling out facial recognition on its Picasa Web Albums site, Google on Tuesday is introducing an updated version of its Picasa software (for Windows | Mac) that can recognize faces in photos stored on users' computers.
Just as it does on the Web, Picasa scans your photos for faces, then groups together photos of specific people. It's then your job to tell it who they are as well as confirm its guesses. If someone you're tagging is in your Google address book, you can also look them up very quickly with auto-complete. Otherwise, Google gives you the option to add them as someone new; this information then gets synced back up your Google address book.
Picasa's software can now scan for faces, and offer up recommendations of people it thinks are your contacts.
(Credit: CNET)The system worked very well for me, but it was slow going. I had to leave the program running overnight for it to finish processing my 3,700 or so photos for faces. It also had my processor humming, since it was doing all the work on my machine instead of Google's giant server farm.
That's not to say Google hasn't included a few things to help speed up the process. For one, if you've got photos that are both hosted online and on your hard drive--and that have already been scanned for faces, the Picasa software can grab that information and add it to your local library. This saves it from having to scan the same photos twice.
And for photos it thinks contain people you've verified as contacts, it gives you quick "yes" and "no" buttons that can add or reject name tags. Oftentimes, clicking "yes" adds a few more suggestions for photos of that person that the program feels is safe enough to recommend. There's also a way to group accept or group decline its suggestions, which saves time you would have otherwise spent clicking the buttons one at a time.
... Read More
You know my stance on bad e-cards, and in the same vein comes my dislike for in-text ad links that you find on some blogs. I'm not talking about Snap's little Web site previews with its Snap Shots service, which people either love or hate, but the IntelliTXT stuff--the kind where you accidently moved your mouse near one and it opens up an ad that doesn't go away for several seconds. Ryan Block from Engadget had a good missive on the matter back in August of last year, and I have to agree with the guy that it ruins the reader experience.
With that said, I'm really digging Viewdle's new Name Widget service, which will cross check any names you mention in a blog post and serve up a tiny little video morsel of the person's face when you hold your cursor over his or her name. The video clips in question come from larger pieces of video that have been run through a facial recognition database and cropped down to fit in an area the size of your thumb. Anyone can add to their blog posts or Web site free of charge with a few lines of JavaScript.
Hovering over the text link of someone's name gives you a quick video clip so you can ID him or her.
(Credit: CNET Networks)The best part is, to actually trigger the video you need to hold your mouse over the link for a good 3 seconds before the video starts playing, so your reading experience won't be too bothered if you make the occasional brush. If you end up actually clicking the name link, Viewdle will kick you over to Reuters, which has a bunch of links to videos where the person appears. Each link jumps you right to that spot. However, Viewdle doesn't require you to link back to its Reuters page, which means you can jump the link wherever you please.
The service works with a variety of popular blogging tools like TypePad, Blogger, and WordPress. However, LiveJournal, Facebook, and MySpace users are out of luck since these sites don't allow JavaScript from outside sites.
You can give it a spin on the names I've added after the break.
Related: Yahoo Shortcuts: It's everywhere you want to blog
... Read More
We're total suckers for video products that let us use our face. A few months back, Caroline put my head on a woman's body with the launch of JibJab's Starring You service, and shortly before that, Rafe played around with Fix8, which lets you overlay live video with avatar and object overlays. Both are vastly different technologies, but the idea is the same: quick entertainment with very little effort on the part of the end user. Along the same lines, Gizmoz, which has been providing 3-D face-mapping technology to the masses since late May, is launching a new product this morning called "be a star."
Be a star is essentially a really simple way to insert yourself into a variety of clips from TV shows, movies, and music videos. Using the same technology found in its basic talking avatar service, Gizmoz will take any photo of someone's face you throw at it, and convert it into a 3-D head that's capable of minute facial movements. You can sort through clips that use just one user face, or multiples that let you add up to three different people. The resulting video can then be stuck on a social networking page, or shared with a simple link.
For now you're limited to 10 clips, but the company is planning to add more every day. Eventually Gizmoz aims to let users use the same technology on any of its videos, allowing you to paste your friend's head on the family dog, or onto the Thanksgiving turkey.
I've embedded a sample video below, which I put together in just a few minutes. The facial processor only hiccupped once, but it was because of my crummy photo of Rafe's head. If you've already got processed 3-D heads in your gallery, it takes a little over a minute to put the entire thing together. My one qualm is that while depending on the original clip, the heads tend to be tiny and the video only comes in one size, although I think you'll agree this is a pretty great way to make embarrassing videos of your boss.
Tonight Gizmoz is launching an upgrade to its service that brings photo face-mapping technology to user avatars. Originally the service allowed users to add voice clips to various pre-existing characters, including celebrities, inanimate objects, and house pets. With this new technology, users can upload a photo of themselves and the service will automatically map their face to a moving, talking character.
Once user faces have been mapped, the service takes on a look and feel a little similar to that of Meez [review] with a build-it-yourself creation tool. Users can adorn their avatar in all sorts of clothing and hairstyle combinations. They can also record 20-second voice clips that will sync up with facial movements and gestures. The result is something that's both visually interesting and a tad creepy (in a Lawnmower Man kind of way).
Gizmoz claims that the entire process takes "under two minutes." Founder and CEO Eyal Gever explained to me that this face-mapping technology was what he originally intended when the service was in its infancy in 2003. It just took a few years to become viable and easy to use.
With the new face-mapping functionality, Gizmoz is also launching two new ways to share user-created talking heads. Previously users could only embed their creation as video clips. Gizmoz is now offering "stickers" and what the service calls an "answering machine." Stickers allow users to record a custom message, almost like a digital Post-it note. This service is aimed mostly for social network comment boards like MySpace.
The new answering machine feature is the more interesting of the two. Users can create and embed their own message (just like Stickers), and visitors can record and respond with their own Gizmoz video responses right inside the player. The new tool reminds me a little bit of Jaxtr's VoiceBlast widget, although a little less businesslike.
Gizmoz is launching these new services in conjunction with a distribution deal with RockYou and Freewebs. Users of both services will now be able to integrate Gizmoz. In my first experience with the site in February, I called it a "gag service." With this update, I think Gizmoz has become a very serious competitor to other social networking add-ons--especially Meez.
I've embedded the answering machine widget to the left. Feel free to add your own responses, and enjoy the head banging.
- prev
- 1
- next
