Posts Tagged ‘imagesearch’

Xerox Working on Algorithm That Can Judge the Aesthetics of Photos

Xerox is showing off a new tool called Aesthetic Image Search over on Open Xerox (the Xerox equivalent of Google Labs). It’s an algorithm being developed at one of the company’s labs that aims to make judging a photograph’s aesthetics something a computer can do.

Many methods for image classification are based on recognition of parts — if you find some wheels and a road, then the picture is more likely to contain a car than a giraffe. But what about quality? What is it about a picture of a building or a flower or a person that makes the image stand out from the hundreds which are taken with a digital camera every day? Here we tackle the difficult task of trying to learn automatically what makes an image special, and makes photo enthusiasts mark it as high quality.

You can play around with a simple demo of the technology here. Don’t tell the Long Beach Police Department about it though — they might use it against photographers.

Aesthetic Image Search (via Quesabesde)

A Simple Bookmarklet for Finding Copyright Infringements of Your Photos

Src Img is an uber-simple bookmarklet created by Jarred Bishop and Hayden Hunter that lets you quickly do a Google Image search for any online photograph with just two clicks. It’s a simple link (i.e. bookmarklet) that you drag into the bookmarks bar of your browser. Whenever you want to search Google Images for a particular photograph, simply click the bookmarklet. It’ll overlay all the photos on the page with a “?¿” square. Click this to search for that photo. Voila!

Src Img (via dvafoto)

Google Adds EXIF Data to Image Search

Google has added EXIF data to Google Images, allowing you to quickly look up details on how a particular photograph was taken (as long as the data is embedded). Simply click any of your search results to see the details in the panel on the right. They don’t seem to be doing anything with geotag info — displaying where the photo was taken on a Google Map, for example — which is probably a smart choice. Something tells me a lot of people would have a problem with that, even though the data is publicly accessible and baked into the photo itself.

(via Google Operating System)


Thanks for the tip, Hannah!

Google Adds TinEye-style Image Search

Bad news for TinEye but good news for photographers: Google is adding reverse image search to their ever-growing list of products, allowing photographers to search using their photos to see the different places they’re being used online. You can search for images by entering a URL, uploading from your computer, dragging and dropping onto the search page, or via Chrome and Firefox extensions. The feature will be rolling out to Google’s users over the next few days — once you see the camera icon in your search bar you’ll know you have it!

Search by Image (via TechCrunch)

Find Photographs that Look Similar To Yours with BYO Image Search

Did you know that Idée Inc., the company behind reverse-image search engine TinEye, also has a web app called BYO Image Search Lab that can take any photo you provide it and find photographs that look similar to it? It’s a neat way to be inspired by how other photographers approached shooting similar scenes.

Imagelogr Causes Uproar Over Photo Indexing Practices, Goes Offline

It looks like Imagelogr, a new search engine for images, has gotten off on the wrong foot. Only weeks old, the service has sparked quite a response from photographers after it became clear that the service was offering copyrighted photographs for download without any links or attribution.

Their “about us” section states,

Imagelogr.com is an image & picture search engine. We try to index pretty much every picture & image currently available on the free internet. With our powerful search engine finding these images should be fairly easy.

The problem was, photographs that weren’t “free” were being indexed as well, including Flickr images marked “All Rights Reserved.”

After learning of the service, notable Flickr photog Thomas Hawk wrote a post on his blog today titled, “Is Imagelogr.com Trying to Be the Largest Copyright Infringer of All Time?“:

Imagelogr claims to be scraping the entire “free web” and seems to have hit Flickr especially hard, copying full-sized images of yours and mine to their own servers where they are hosting them without any attribution or links back to the original image in violation of all available licenses on Flickr. If people on Imagelogr want to they can manipulate your images, rotate them, see them at different sizes up to 300% and even download the images with a download button directly from the site. […] The site currently boasts to be tracking over 24 *billion* (yes, billion with a B) images. If their numbers are true, this may in fact be the largest image grab in the history of Flickr.

A thread about the service was also created in the Flickr forums, but was quickly closed by the Flickr staff, since they felt that the thread was going down “Lynch Mob Road”.

Hours later, the service was taken offline, and was replaced with a simple message stating,

Imagelogr.com is currently offline as we are improving the website. Due to copyright issues we are now changing some stuff around to make people happy. Please check back soon.

We don’t know much about this service, and hadn’t heard of it prior to today’s events. Their domain name was registered last month, and details are so non-existant that it feels almost like a class project. However, the fact that they’ve indexed 24 billion photos seems to argue against that possibility.

When they come back online, you might want to type in your Flickr username to see if your photographs show up.


Image credit: Screenshot by Thomas Hawk

NachoFoto Launches Real Time Photo Search Engine

NachoFoto is a new image search engine that attempts to deliver relevant results for a specific kind of query traditional search engines haven’t focused on: dynamic keywords.

These are keywords for which the resulting photographs should change over time.

For example, if someone searched for “tiger woods” this past week, they were likely looking for photographs of him at the Masters. However, traditional search engines such as Google returned exactly the same images as they did the week before. A quick Google search for “Tiger Woods” shows many images of him, but nothing specifically from this week. A NachoFoto search of the same term returns photographs ordered by freshness.

Another example would be the searches for “earthquake”. Those who search for the term “earthquake” prior to a major disaster would have greatly different expectations than those who searched for the term immediately after. As of now, Google does not offer any way to sort or filter by time in their image search.

Traditional services like Google built their reputation upon relevancy, but newer services such as Twitter have demonstrated that the ability to surface “trending” topics is important to users as well.

This new service is an interesting look at a feature image search engines should have, but unless someone acquires NachoFoto, it probably won’t stand a chance if the feature is added to existing search engines.

(via TechCrunch)