Want to know more about Tagcow’s consumer services and how we can help increase the value of your consumer applications like data backup services, photo sharing and photo print services (calendars, coffee mugs, post cards, prints, etc.,)? Remember, you have the images but without meta data you don’t have an asset! Image Data Equation: assets = your image + Tagcow meta data.

Meet our CEO, David Cantu, at CES Las Vegas from January 6th - 8th.

To setup a 1:1 meeting with David call him at 425-894-2155.

Event: http://www.cesweb.org/

Rihanna, and it wasn’t even close. Rihanna’s image was tagged almost 2x as much as the second most popular celebrity in ‘09, Lady GaGa.

From Tagcow

Coming off the heals of another celebrity tagging job I thought it would be interesting to take a different look at the data. So I set out to answer a couple of simple questions. First, what is the typical “mood” of a celebrity in the photos taken? Second, are there more female or male celebrities photographed? I’ll answer the question of “Most” photographed celebrity of 2009 in a special New Year’s Eve post tomorrow.

Celebrity Moods
As you can see most the overwhelming mood of photographed celebrities is Happy - 59% and Sexy - 25%. I guess we’d all feel happy and sexy if we were rich and famous. Chart below…

From

Celebrity Photo Gender Distribution
As you can see female celebrities have their photos taken roughly 12% more often than their male counterparts. The only surprise here is that there’s not a bigger discrepancy. I blame the Twilight Phenomenon of 2009. I think we tagged a million images of Robert Patinson and Taylor Lautner this year.

From Tagcow

As Chief Business Architect the best part of my job is that I have a lot of room to run out in front of the execution / operation team to spend money and to do things that the company otherwise wouldn’t do. I recently hired a new developer to the team, Semi Mol, who will be assisting me in a new “Skunk Works” project. While I obviously can’t get into what exactly we’ll be collaborating on I can tell you that it’s going back to the core of the Tagcow dream. When Matt and I conceived the idea for Tagcow over Sushi in December of 2007 we had one big idea - searching your personal image archive would be cool. Since then we’ve drifted away from the consumer image tagging market and moved towards the Enterprise market because getting a few companies to write big checks is easier than getting millions of consumers to write you small checks.

While we’ve been off busy executing on our Enterprise strategy I kind of figured that with all the noise being generated out of Facebook, Flickr, Photobucket and others that somehow this problem was well on its way to being solved but then I looked a little closer…and you know what? All the noise was really around “People Tagging” either manually or through facial recognition software and honestly it’s not that interesting. It’s fundamental but not game changing. I’d recommend Face.com’s facial recognition software any day and if I had some spare time I’d call those guys and integrate their software into our platform to help with people tagging. But the real heart of the tagging opportunity lies with “descriptive tagging” of images which no one does better than Tagcow. So you have 400 photos of your daughter, Novella, tagged now what? It’s the ability to narrow down your results to “Novella + ballet” or “Novella + laughing” that’s really interesting.

As Semi and I make progress on this front I will keep the blog updated.

Welcome to the team Semi, it’s going to be a wild ride

Semi Mol

Okay, so we didn’t actually run it but we did tag all the images from the marathon. Very cool!

Statistics from the photo tagging job below.

Stats:
Est. Runners: 2,500
Total Images: 2,379
Unique Runners Identified: 1,057 (42%)
Total Tags: 4,739

Total cost: $47.58
Est. Time: 10 minutes

And by the way, I was completely wrong about the emotional tags for the London Santa Marathon it looked like the runners had a tone of fun! This job is a milestone for Tagcow as it represents our first attempt at event tagging. I’d say a smashing success. I am sure Brightroom will be calling us any moment to start tagging their events.

Event details.

To see some fun photos from the event go to 42run.com and search Santa Run 6k 2009 using the following bib numbers:
990
849
1425
1808

Who said working on Saturday isn’t fun…

I was just notified that our consumer image tagging service was featured on the popular podcast, Faceoff Show. They do a fair job of describing how consumers can get their photos tagged using Tagcow for a small fee and what the benefits are. They also discuss our brilliant use of Mechanical Turk and proprietary algorithms to solve one of the most complex computer science problems. I’ve always liked their show and now of course I love it!!

The segment on Tagcow’s image tagging service runs from the 11:43 minute mark to 18:57 - over seven minutes…super rad!

Faceoff is your face-to-face web technology podcast. In this podcast Jade Robbins and Mark Sanborn talk about various aspects of web technology such as web development, social media, and web entrepreneurship.

Listen here

Thanks guys!
-Michael Droz

We’ve been photo tagging celebrities recently so I though it’d be fun to post a tag cloud from the recent batch of images. Apparently Rihanna, Pattinson and someone named Gaga are popular these days - who would have know? I don’t see Madonna or Clinton anywhere…what decade is it anyway?

We’re going to continue to post these tag clouds from our photo tagging service every couple of days so keep posted they can be a lot of fun. I think I will do one for emotional tags and then maybe one from a marathon that we are going to be tagging.

I’d bet a million dollars that the number one emotional tag from the marathon will be “miserable” - but that’s just my opinion. Don’t take offense my running friends you know I say that out of respect. I can run fast but not very far

Finally!! We’ve released version 2.0 of  Tagcow with a brand new tagging template for our consumer users that seriously kicks butt! The new template really pushes our image tagging system to identify unique objects, create minimum tag sets and focus on the emotional aspects of the image as it’s being tagged.  The results of the update trump any automatic object recognition system by decades.

Checkout the tags on this simple photo of me and Anakin…

unique_attributes:
Watching|Happiness|Thinking|Sitting
main_theme:
Boy and man
image_type:
photo
color_type:
color
key_objects:
Boy|Man|Drums|

and a simple pic of me playing vids with Anakin produced these tags:

main_theme:
Playing video game
color_type:
color
unique_attributes:
concentration|relaxation|serious|family
key_objects:
Man|Young Boy|Couch|Game controllers
image_type:
photo

We are getting very good and consistent results from the new template. The best news is that with these improvements we’ve actually lowered the cost of our consumer description tagging service from 2.5 cents per image to 2 cents per image and we still offer a money back guarantee if you’re not satisfied with the results of our service. What we’re working on next for our consumer service are more custom controls allowing users to control their tagging to the very last detail. Including whether or not to use quotes to separate tags or to include common misspellings and stemming in your tagging results.

Don’t count Tagcow out! As a very prominant venture partner recently said, someone is going to solve the image tagging problem for consumers and it’s most likely going to be Tagcow. Umm…mostly likely?

It is commonly known that search engines are currently incapable to identifying objects or content within the image and although there has been much work recently for computers to perform this work, the technology is not quite there. This problem leads to one of the most common questions we see pertaining to the value of image or photo tagging.  The answer to this question depends on the solution you are seeking. General users with personal image collections have different needs than those of enterprise customers but the problem (and solution) is the same.

Personal

Personal or power users typically have growing collections of images that are stored in a flat directory structure.  Unless they are disciplined or exact about how those images are stored and cataloged, images are stored by date and maybe by event.  However, most users don’t go so far as to change the image naming structure so even when they are organized by date, there is no detail on what is in the photo.  What they end up with is a bunch of images with file names like “DSC001248.jpg.”  Not very descriptive to say the least!  What this means to the user is if they want to find all pictures of their daughter, they must rely on their memory or spend time looking through thumbnail images to find images they are searching for.  While this may be a viable solution, it is not very practical as their collection grows from 100’s to 1000’s of digital images over time.

Enterprise

Enterprise customers have a similar problem except theirs can cost them money.  Enterprises have large collections of images that are stored in repositories.  Although they tend to do a better job at renaming their images with names that relate to the content in the image, it is not practical to list out every detail in the image and include that in the file name.  Alternatively, some enterprise customers will create elaborate directory structures to account for the insufficiency in identifying image content.  However, as the repositories grow, enterprise customers are faced with an ever growing collection of images that are essentially lost or undiscovered because of the shear number of images in the repository.  This can cost a company sales especially if a customer is unable to find a particular piece of art, for example.

So what can they do.  Tagcow was started to address this issue.  Instead of waiting for the technology to arrive to discover the content within an image, we approached the problem from a human point of view.  Humans are uniquely qualified to identify the content in an image and will, for the foreseeable future, be able to add tags, such as human emotions, that computers will find difficult to ever tag.  That said, image tags will empower users, personal and enterprise, to search their images or photos in a way never before possible: by content.

Prior to changing our service over to a pay for play model we conducted a user survey of our free service and the results were very positive. They indicate that our users were generally excited, optimistic and satisfied with our service.  Using these data points we confidently started charging for our service. Not that we had a choice because the operational cost of the free service were killing us. The thing is once we made the switch to a pay model our users’ perception of quality made a 180 degree turn. Based on the declining daily registration numbers and negative feedback we got from our user base it was obvious that quality is a function of perception. When users didn’t have to pay they really liked our quality when we asked them to shell out for it they had a very different opinion. The good news is that it’s clear that if we improve the quality of our service users will pay.

Below are some of the highlights of that user survey: