image search
EBay adding image recognition to mobile shopping apps
USA Today reports that EBay is planning to add a shopping by image recognition capability to its mobile apps. Customers will be able to send a picture of an item to EBay, which will match it to similar items that can be purchased on Ebay.
Related articles:
New Kooaba release provides significantly faster image recognition times
Computer vision startup Kooaba reports in its blog that the new release of its recognition technology is about 40% faster on average, taking 2.5 seconds, including the roundtrip to the cloud. The speedup applies to all of its products, including object identification, news/magazine article identification, and its general visual recognition API.
Visual search shopping technology moving to Google Product Search
The computer vision technology enabling online shopping by visual similarity in Boutiques.com will be moved to Google Product Search, according to an official Google blog post written by Burak Gokturk. Gokturk was a co-founder of Like.com, which Google acquired last year for $100 million to build Boutiques.com. Miranda Miller of Search Engine Watch reports that Boutiques.com's traffic had fallen by 94% by April of 2011, to just 170,000 visitors per month.
Related articles:
VideoSurf releases its mobile TV show recognition app on iPhone
In a recent press release, computer vision startup VideoSurf announced that it has released its mobile application for iPhone. The same application is also available on Android platforms. The application can be used to recognize TV shows and content from other video sources. "Users can discover videos by simply pointing their phones at a television or computer screen to capture a few seconds of what they are watching." Information on the show and cast, unique content, similar videos to watch can be provided by the mobile application. There is also a social-TV component that allows users to share details on Facebook.
Related content on Computer Vision Central:
Picitup announces cloud-based suite of visual similary shopping tools
Picitup is releasing its PicCommerce visual search suite which will enable any e-commerce site to have a visually based recommendation engine. Until now, Picitup's visual search technology was used only by large online retailers such as eBay and Shopping.com. PicCommerce provides capabilities for users to shop by color and by visual similarity. The software runs on the cloud so that local installation is not necessary. More information is available in a press release.
Related articles:
Image search engines compromised with poisoned images
According to Trend Micro, hackers have been able to fool image search engines by "poisoning the images". The hackers would monitor Google Trends for hot keywords, and then automatically generate material in new Web pages that would give fake results. As search engine bots roam the Web, the fake content would be chosen since it is related to a trending topic, and consequently, it would appear near the top of relevant search results. The latest scam involved poisoning the image search such that a user would click on thumbnails of search results and get redirected to a fake antivirus website, encouraging the user to buy unnecessary antivirus software. More information is available in a NewScientist article.
Getty partners with Google to provide rich museum experience
The J. Paul Getty Museum announced that it is the first museum to make its entire collection of paintings available via Google Goggles. With the Getty version of the Google Goggles app, visitors can point a smartphone camera at a painting to access information about that painting or its artist: much more information than a curator can place on a traditional wall plaque. According to the press release, the computer vision algorithms underlying the app are able to discriminate between different versions of the same painting.
Related article on Computer Vision Central:
Pongr announces visual search engine for measuring brand sentiment
Computer vision startup Pongr announced its ImagePulse visual search engine for measuring brand sentiment. It recognizes company logos in images that people post on twitter and Facebook, in order to measure brand popularity. Pongr's computer vision technology enables cross-linking of brands to identify clusters of products purchased by certain demographics. More information can be found in this press release and on the Pongr website.
Related articles on Computer Vision Central:
Google rolling out content-based image search
Google is rolling out a new content-based image retrieval (CBIR) capability at images.google.com. You can drag an image from your desktop to search for similar images, or you can enter the URL of the image if it is on the web. With the Firefox or Chrome browser extensions, you can search directly by simply clicking on an image in a web page. Google's image search is being released over several days: it is available to you if you see a camera icon in the Google search bar.
VideoSurf app ID's videos and TV shows
Computer vision startup VideoSurf announced a new mobile app for Android smartphones with a V-ID feature. The user captures a few seconds of a playing video or TV show with the smartphone camera, and the VideoSurf app identifies the program. The app also allows users to post comments about the show on Facebook and Twitter. The app is available at www.videosurf.com/mobile and through Verizon's VCAST apps.
Related articles on Computer Vision Central: