June 17, 2008 by pmelcher.

- A useful blog. With some delay but with quite a bang, Photoshelter finally launches a very useful blog for its users and beyond. Full of tips, info, rules and dynamism, it could very well become a very helpful resources for stock photographers who take photography seriously. Does that mean they plan to close the other useless egg hugging blog who rips off hundreds of valuable images for free under the cover of ‘fair use” ? Go ahead, shoot that blog

more on School of Stock here:

- Multimedia continues to rule : Ed Kashi  has launched a wonderful website entirely dedicated to his work on Nigeria delta. Curse of the black gold, offers, among other option, a great multimedia who has all the attributes of a Mediastorm production. A must see, keeping in mind that Kashi was briefly captured and jailed to bring this issue to the world.

Multimedia here:

Update : June 19, 2008 : this article on MSNBC :

Nigerian oil field shut after U.S. worker seized

- Geolocation without GPS: Geolocation is the ability to pinpoint the location, on a map, of where an image was taken. Carnegie Mellon University took millions of  already geographically tagged  images from Flickr as a tool to identify the location of an image, any image. Works a bit like this. You upload an image which compared to millions of Flickr set. By recognizing attributes, it can almost accurately find out, by itself, where the image was taken. Using the same principle, one can easily see how automated keywording could benefit from this crowdsourcing approach. More details here

- Getty’s latest set of numbers: Funny how no one noticed how the Wireimage brand took a huge beating after being purchased by Getty. According to Getty’s published number, Wireimage went from  + $ 3 million a quarter, to a few hundred thousands the next full quarters. What happen ? Did the Wireimage staff just stopped working ? Furthermore, PumpAudio seemed to have also fallen to zero revenue for two quarters after acquisition.

Posted in mediastorm, celebrity, technology, commercial stock, multimedia, Search, slideshow, flickr, web 2.0, keyword, getty | Print | No Comments »

June 10, 2008 by pmelcher.

We are getting there..slowly.  It is not an easy road, but we are getting closer.  In the last year or so, we have seen more and more image search companies come out and expose themselves. Even the even mightiest, and certainly the worst, Google Image is thinking about changing its algorithm.

The holy grail is, of course, the end of the keyword based search ( aaaargh !). The first baby step we are currently seeing only focuses on face recognition. For two main reasons :

- A face is always a face, a triangle between two eyes and a mouth, and rather easy for a computer to recognize.

- Between celebrities and relatives, when you deal with image search, the majority of people are looking for either friends, relatives, themselves or more pix of celebrities . There is a huge market.

So here goes many worldwide software engineers claiming image search nirvana. At least when looking for people. I have tested a few, recently, all in Beta and none quite there yet.

TinEye: In a league of it own right now as it does much more than just face recognition image search. Probably the most advanced of all,  its limitation is its extreme accuracy. Looking for an image and it will find that image, nothing less, nothing more. Every altered version of it. Great for many, many usages, but a bit limited for those just seeking a similar or inspiration.

Very far away are :

- Polar Rose. Swedish based, it has been full of promises for many years but with disappointing result up to now. Also starting in the face recognition, its algorithm just became clear to me when it recently invited me to Beta test its Plug in. I was expecting a TineEye sort of plug in, but instead, they put me to work. What is up with that ? .

The plug in works like this : every time you are on a web page, it scans every image for a face. It then puts a square around the face and asks you to put the name. As you enter the name, it starts suggesting options. It was always right on the money. At first, I was really impressed, as in ” How does it know ?”. And briefly later on, I recognized the trick. It scan any available text around the image, looks for two words next to each other starting with a capital letter and assumes that should be the name of the person in the pic. All I have to do is confirm. Thus, Polar Rose is currently no more than an elaborate and free version of Amazon’s Mechanical Turk. Putting the community to work. While there nothing wrong, in principal, with this approach, with all these website these days that ask for my free input, I am close to putting full 24 hours days of work, for free.

Their chances of success, using that approach, is minimal. I, for one, will not become a slave to the machine.

- Picollator : Russian by birth, this new image search also looks for faces. You can even upload a sketch and it will try to match it. I tried with this image :

surprisingly got no result.  So I went for the TinEye favorite, the Mona Lisa:

.

I got this :

Some matching results, some very weird ones.

Face recognition and matching can be a very useful tool for the news and celebrity world as photographers and editors could quickly edit film with proper name spelling by scanning the web for matching results. Especially for those B to D celebrities. It is, at this stage, still very sketchy and not quite ready for the big time.

It is also a good indication of how far we still are from a computer recognizing, properly, everything in an image. Right now, we are only scratching the matching part and even that, has it flaws.

It is also very important to know that all these sites are in Beta, meaning far from claiming 100 % accuracy and should not be dismissed until officially proven ineffective. You can try all of them, for yourself :

Tineye

Polar Rose

Picallotaor

Posted in Canada, technology, idee, celebrity, keyword, web 2.0 | Print | 1 Comment »

May 27, 2008 by pmelcher.

Sometimes it is good to step out in order to better look in. In this 25 minutes video, Peter Norvig, Director of Research at Google, goes over image search, segmentation and other challenges that face anyone that is trying to built a better search.  What is interesting in this is how he speaks over and over about that its all about working with the data you have and not power coding. If you analyze the data you have ( data can be replaced by images, obviously) then you get a more agile platform.

In the case of photography, we tend to add more data ( think keywords) to an already huge set of information. We wrap data with more data, which never made sense. Google, TinEye, Riya and many other companies seem to have taken the lead in a area where photo agencies should have been the precursors. It is fascinating to see how many companies will sell you a Image Database management system that is really a keyword database management with images attached to them. None even use advance text search technology and let you suffer thought metada building on your own.

On the flip side, Image Buyers are still relying on a guessing game, trying to find the right image with a varied source of words. Very archaic, isn’t it?

Take a peak at the video here :

Posted in technology, Search, keyword, google, web 2.0 | Print | No Comments »

May 4, 2008 by pmelcher.

“The big winner in these data are Internet design and development firms, whose purchases dominate the industry. But while these firms consistently hold the highest market share, both in terms of dollars spent and units purchased, their average spending per image was often lower than that for other segments, particularly creatives. These dynamics can have strong implications for producers and marketers of stock imagery.”

TrendWatch Graphic Arts_2005_

That was 3 years ago. The $71.9 million revenue of Istockphoto in 2007 is a clear proof that this study was correct. It actually did not reveal any secret but confirmed what everyone already knew. So, how as the industry reacted in the last 3 years ?

- The conservatives: Most have continued in their all too familiar ways, producing the same images for the same pricing, even when 90% are now being sold at microstock prices. Voluntary ignoring the microstock noise, that has now grown from a chatter to a scream, they believe that it is just a passing fluke that will die down when the party is over. Some even take a stand against this market by claiming loud and clear that they will NEVER surrender to such pricing model. As if high-priced images had any appeal to any image buyers.  They have the same images available on mid and micro sites, yet priced according to antiquated models. They have capture the attention of the medium mediocre pro photographers whose very livelihood is threatened by this sudden price drop. As the protectors of the pricing “status quo” they stand up desperately like little toy store owners in front of K Mart refusing to sell cheap chinese toys while charging 6 times more for the same products.

- The cold feeters : They understand what the market is up too but have a hard time accepting it. They have been in this industry for a long time with reasonable success and hate seeing prices go down. So they adapt. They either create new collections for these markets or “retire” images and price them lower. One foot in the water, one on land, they think they can continue doing what they know best without loosing on new pricing trends. They enter with little careful steps as walking in the waters of a cold lake, trying to organically figure out its temperature and if they will be able to swim. These agencies confuse their photographers tremendously as revenues become unpredictable.

- The passionate: They can’t get enough. They were early-adopters and cannot wait to do more. Some have caught the wave at the right time and are seeing good results, while others waited and jumped both feet at the same time. They are leaving familiar pricing territories and well know clients for hit and run sales. Less tear sheets but more volume. Less marketing intelligence but more sales data. Some have created their own micro/mid platforms and are hoping to catch up on the Istock wave ( Tsunami ?).

What is fascinating is when you see a traditional RM/RF platform asking their contributors for more “regular business images: people at their desk, answering phones..” . It is also laughable when others stand up on their soap box and say: “The answer is easy, my friends, create more high end RM images”.

Huh ?

When I was at Corbis, the term was “cutting edge” and even back then I was hitting my head against the walls trying to figure out what it meant. And more important, how it is made.

A stock image is by nature an image that is made to please a wide variety of clients. Furthermore, it has to be unfinished so that text and logos can be added. Even other images copied and past into it. Even with our world turning into a global village, local sensitivities are at skin edge. It makes it very hard to create images that will please everyone in the world.

So what is high end RM images ? I would like to see. Because, like the Eldorado or Atlantis before, there is a lot of talk about it but not much to see : Behold the savior of photography, the almighty cutting edge, high end photograph. The secret knowledge that will lead to the golden caves of fortune: How to create a high end image. If you possess that wisdom, that elusive stone, the magic wang that transforms a low end / medium end image into that legendary 6 figure image, then why in the hell would you stick it in a photo agency that will take 50% or more of your sales ?

Most commercial stock photo agencies, right now, should give their salespeople a substantial raise. Because the major reason their sales are still stable, if not rising, is because their client stick with them. Mostly because they enjoy the help and friendship with their contacts there . Its not the pricing, nor the images, those can quite frankly be found elsewhere, its that person on the other side of the phone. Yet, most go on a search for the elusive “sharp, cutting edge, high end” imagery soon to be copied by extremely web savvy microstock shooter.

A good salesperson will tell you exactly what images you need because they sell them everyday. Its not going to be a creative research ( shoot RV people !!, especially the ones being pushed because people can’t afford the gas price, says Corbis ) or keyword search analysis. Not even past sales data. It is going to be these guys behind their screens and desks, day in and day out that can tell you what you need. And they also, have no idea what a cutting edge image is, but they sure do know what an image that sell can be.

Posted in Search, Midstock, license, commercial stock, keyword, finance, Royalty free, corbis, transaction, Microstock | Print | No Comments »

April 29, 2008 by pmelcher.

Google is thinking about changing its Image search algorithm. Currently it has a convoluted way to return results. As you probably already know, its a basic “text” search which looks at the file name, “alt” comments and words around the image to declare an image a valid candidate to a search. Meaning that if you search for “cat” for example and someone has named an image of a truck “cat.jpg”, has put the description in the “alt” comments as “cat, cat and more cat” all that around an article about how great his cat is, then that image of the truck will appear in your search.

Not very efficient, is it ? And as previously written here, the image could be completly out of focus and grainy, as long as it meets all the requirement for Pagerank, it will appear high up in the results.

Called “PageRank for Product Image Search” and presented at the International World Wide Web Conference in Beijing by two Google staff engineers, it is aim at becoming the new VisualRank.

Claiming to be an image recognition system and using advance object recognition, here is what it does. It scans all images and looks for patterns, regardless of what object is in the image. After a while, it will see that some images or at least part of the image have the same pattern . Those will be linked. The image or images that have the most similarities with all the others will be pushed to the surface.It gives you a result like that :

See that image in the middle ? It contains all the attributes of the others, thus its the most relevant.

If applied this will create a headache for the photo business. Since this search is really made to search for products to purchase and not for images to license, it is counter productive for our industry. It will not return the best image, the most liked or the most striking, only the most banal, the most common. Ouch !

It will favor non exclusive images, think RF and microstock, over RM images. It will enhance the most used images not the best ones. It will slowly bring IPTC kewording to obsolescence.

In order to bring traffic to its website, a photo agency or photographer will have to post  images as much as possible everywhere all the time.The same image. Thus an image with a lot of various usage will be the star, while news images, who usually have a shorter life-span, will not score well. But an image of a spoon might become a superstar. Especially if it is sold everywhere

Google hates photography. Or rather it sees  it as a tool, not as an art. Another way to index the world.It will become harder to find great images with Google and that will continue to open a door wide open for anyone looking to create a search engine for photography with a ranking system based on quality and relevancy. An image search and not a visual search.

More on the emmerging proposal at Techcrunch.

Posted in IPTC, Search, technology, keyword, google, filter, flickr, editorial | Print | No Comments »

April 24, 2008 by pmelcher.

Image details: Pope Benedict XVI Celebrates Mass At Yankee Stadium served by picapp.com

It is a blessing these days to see a company that grows by listening to its criticism. Ad supported licensing company Picapp has recently revamped its site and has made some good improvements.  First and foremost, you can decide if you want or not that little pop up figure they call picaboo. You can also choose the size of the image you want to post and what type of an animation.

Thus after selecting your image, you can select between a goofy interface to a more serious one. What the people needed.

A quick down and dirty Alexa ranking shows Picapp  ahead of competitor Gumgum in traffic. No big surprise as they started with a big bang using the Getty  trampoline.

The real question is why did neither of these companies have open their service to individuals.

Sure, it is nice to have access to images from pros, but what about the huge pool of amateurs. This licensing model would much better serve the Flickr community than anyone else. After all, it would be a great replacement to the useless Creative Common scheme. ” here, use my picture for free, in exchange for which I get a cut on ads”. Fair enough, no ?

But neither Flickr, Photobucket, Smugmug and other mass photo storage platform will allow their content to be duplicate on either the Picapp or Gumgum server. The technology has to come to them. And that is the biggest shortcoming of both companies. Because their technology is neither proprietary neither that hard to create.

The second short coming is that neither offer the publisher any income for posting these images. Even the slightest cut would make either company immediately attractive. Imagine, get paid to post images !! Someone is bound to do it.

I can foresee very soon many  agencies offering the same type of licensing model from within their own site, bypassing the “Picgum” middle man.It would not be a problem for Flickr to add that option too. So it leaves both companies in breathless race to create enough critical mass of content to become indispensable. One, Picapp, has concentrate on overall volume, while the other, GumGum, seems to concentrate on just celebrity oriented content.

Let’s see what the future brings them. Either way, a very interesting race to watch.

Photo Licensing by GumGum | © PacificCoastNews

Posted in celebrity, copyright, alexa, technology, gumgum, license, keyword, filter, editorial, flickr, prosumer, web 2.0, getty | Print | No Comments »

April 16, 2008 by pmelcher.

I did not pick this image. I actually have no idea what it will be before I publish this entry. Why ? because it is a sort of semi “intelligent” algorithm in the background that will do it for me. A bit like Google ads scans a whole web page for keywords and post the relevant ads, this system, delivered by Dailylife.com, does the same.

It will scan the page for keywords and post the most appropriate image. Like an automated photo editor. And because it is looking thought the feeds of Reuters, GettyImages and AP, I believe, it selects from a pool of already very tightly edited images. One could also foresee a Flickr API, a bit like I did with the yahoo pipes.

I am guaranteed a good and hopefully, relevant image . This is the future of news photo editing on the web. At least for sites that do not care so much about the image and use them as an illustration of  a written report. Why pay some guy to look at a stream of pre edited  images, download one, resize it and post when the whole thing can be automated. And better yet, computers don’t whine, do not take lunch breaks, or holidays and never, never ask for a raise. So why keep a web photo editor, if only to do some “best of the week” gallery ?

Think about it:  the biggest news source of the internet has no photo editor. It is called Google news and it selects images with a similar technology. Indeed, it relies on images previously edited by pro photo editors. For now.

The dailylife link is completely free, with no uncontrolled ads, like a Picapp or a GumGum would like you to swallow. Sure , it has a link for the site itself but the same technology could easily be applied by anyone on their own site.

Finally, Dailylife.com, still in Beta, looks like an interesting destination. It seems they want to be a new Google news but put a heavy emphasis on photography and has a much better and smoother interface. More like a magazine designed for the internet, and not the opposite. Finally.

As newspapers and magazine are suffering more layouts as ad spending is weakening, most of the photo related professional are turning to the internet. However, because of its built in automation, it just seems that some of the jobs will not be recycle but ultimately replaced by machines. We will still need great pictures, thus talented photographers. Not so sure about needing photo editors.

Posted in multimedia, Search, newspaper, digg, magazine, gumgum, technology, keyword, yahoo, editorial, news, slideshow, wire service, google, photojournalism, getty | Print | 2 Comments »

April 14, 2008 by pmelcher.

 Ever since I wrote the piece “orphan work is good for you”, I have been slammed with emails which all pretty much revolved around the same issue : Orphan work bill is a open door for Copyright Infringement and the legalization of free usage.

First and foremost, I keep on refering my readers to a very concise and readable link created by the Copyright Office . In no way do they propose or would facilitate free usage. It is not their intend. It is also quite admirable that they have, and will continue to hold hearings  to listen to all and everyones concerns.

But the most important, to me, is to see that Canada has had a Orphan Work bill in effect since 2005. Quite effective, it requires the potential user of an Orphan image, after proving its good faith in researching the copyright owner,  to be granted a license by the Copyright office. It has happened  only 19 since  then. Granted, Canada is not as big as the United States, but 19 is almost not worth a second look.

Furthermore, the Canadian law leaves 5 years after the license has been granted for the owner to retrieve his/her license. It is still early to see if this provision has ever been used by anyone.

The copyright office, and any all serious image licensor know very well that technology can be a very serious ally in copyright protection. Regardless of any petition and whining, the bill WILL pass.

As said before, It will be a great  incentive for everyone to respect metadata more seriously  and for the photo industry to finally grow up.

Posted in idee, Canada, focus, technology, copyright, license, web 2.0, keyword, IPTC, law | Print | No Comments »

April 2, 2008 by pmelcher.

A new company emerged from unknown depth a few days ago, proposing free “automated” tagging, or keywording. Named Tagcow, the company does not explain how the tagging is done.

Curious, I decided to give it a spin with a couple of images including this image:

After two days, the image was finally tagged with two words: “Pool” , “Man”. I guess that is what you get for free. Furthermore, when I downloaded the image, I could not find the keywords anymore…There is nothing automated about this service. It is currently impossible for a computer to recognize  the content of an image. The most advanced systems I have seen have a 10% success rate, and then again with very contrasted and simple pictures. Tagcow uses a little known service offered by Amazon call the Mechanical Turk. With this service, anyone can put a long, painful task and offer to pay for human beings to fulfill them

“Complete simple tasks that people do better than computers. And, get paid for it”. At 0.01 cent a picture, one can get images keyworded for cheap. However, the quality is not guaranteed.

Thus it is the power of the masses used here, making more obvious why they picked a cow for their name. Not the brightest animal in the land

Posted in No sense, Search, copyright, IPTC, keyword, flickr, prosumer, web 2.0, filter | Print | 2 Comments »

April 1, 2008 by pmelcher.

How many times will  members of this industry get together and talk about IPTC, keywording and other metadata. How many meetings, conferences, synopsis, “get togethers”, panels, parties, does it take ? Both the ASPP conference in Arizona and CEPIC in Malta have scheduled hours long conferences on this subject. Again. The one in CEPIC is 8 hours long !!!

It used to be that the IPTC was a small geeky association of nerds looking into how to standardize metadata in images. It has now become the most sought-after organization. More than the dying PLUS coalition.

The amusing part is that none of the attendants are keywording their own images. They have staff people to do that. Furthermore, none of the companies that offer this service are on the panels( JaincoTech, Keedup, OnAsia Digital, Etc) They would know better, wouldn’t they?  Instead, you have marketing managers or agencies owners sitting in stuffy rooms, vaguely writing notes while waiting for the suffering to end until they can finally get a free drink at the evening’s cocktail party.

At a time when the temple of controlled vocabulary  ( the Library of Congress) has decided to pull out from its antiquated method of keywording by putting 4,000 of its images into Flickr and ask for crowdsourcing wisdom, the photo agency world is wasting time and money into desperately trying to impose a standardized form of controlled vocabulary. Some probably spend more time and money on attending these panels than they do in a whole year of marketing.

The aim, apparently, is to define a series of code words that could be transported from one databank to another and yield the same results. Thousands of them. Same keywords. Wether they are related to the real world is irrelevant as the priority is to standardize and eventually give photo buyers a book on which word to use and how. And then what, have university offer a degree in photo researching, transforming photo buyers into bonified librarians? It is bad enough that some “photo editors” have no visual experience, it gives me the shivers to think what would become if this would happen.

There are few misconceptions here :

Language, unless dead, evolves all the time. Even dictionary publishers worldwide know  as they add and delete words every year.  Who uses “walkman” anymore ?

Keywording is not a marketing tool:  A bad or irrelevant  image well keyworded is still a bad image. It will not sell.

Controlled Vocabulary does not include local cultures. If it does, than it cannot be controlled anymore. It is arrogant, pedant and quite simply foolish to even believe that one controlled vocabulary can and will apply to the whole world.

A word is not a definition. It is only a description. It takes many words to skim the surface of what an image is. Thus keywording should be an accessory to search, not its main engine.

In the long term, keywording will die. Already, there are other emerging ways to search for images : visual, color, face recognition, similar, pattern recognition. In the text world, there is even semantic search, which allows you to search by meaning instead of exact match.

Google images, which everyone sees as the ultimate “find me tool” does not even index IPTC.

They say insanity is repeating the same thing over and over hoping for a different result. Seems to be that the photo industry is banging over and over on the same door and it will just not open.

The solutions ? Exactly what the user generated content agencies are doing. Let the keywords be offered by the source . They shoot, they keyword. And they keyword well because they are using an everyday vocabulary that the buyers are also using. A vocabulary that changes and evolves all the time. A vocabulary that is not “controlled”. Organized chaos.

Or follow the giants. Getty, which you never see at these repeated panels, as well as AP, Reuters, Corbis and others, have hired outside companies to do their keywords. Because it is not their chore business and do not feel it necessary to have a full time dedicated team of librarians. They seem to prefer selling images rather than cataloguing them.

It would be an interesting exercise to calculate how much time was wasted in “perfectly” keywording images that never sold in some of the medium or smaller agencies that seem to be obsessive about doing in house keywording .

Would it be more interesting for these congress, meetings, conference to have a panel about how to make great pictures that sale ?

Worst that could happen would be a few hours looking at great image.

Posted in keyword, Search, Plus, Zymmetrical, google, flickr, corbis, PACA, CEPIC, getty | Print | No Comments »