Friday, October 2, 2009

Image Retrieval Systems

Untitled - 1

MIRROR (MPEG-7 Image Retrieval Refinement based On Relevance feedback) is a platform for content-based image retrieval (CBIR) research and development using MPEG-7 technologies.

    MIRROR supports several MPEG-7 visual descriptors:

    • Color Descriptors:
      • Dominant Color Descriptor (DCD)
      • Scalable Color Descriptor (SCD)
      • Color Layout Descriptor (CLD)
      • Color Structure Descriptor (CSD)
    • Texture Descriptors:
      • Edge Historgram Discriptor (EHD)
      • Homogeneous Texture Descriptor (HTD)

The system core is based on MPEG-7 Experimentation Mode (XM) with web-based user interface for query by image example retrieval. A new Merged Color Palette approach for DCD similarity measure and relevance feedback are also developed in this system. The system is highly modularized, new algorithms, new ground truth set, and even new image database can be added easily.

2.IBM Research - Intelligent Information Management Department
The Intelligent Information Management Department at the IBM T. J. Watson Research Center is addressing technical challenges in database systems and information management.  The department includes the Database Research Group and Intelligent Information Analysis Group.  The department exploring novel techniques for indexing, analyzing, fusing, searching and exploiting structured data and unstructured information in various scientific and business contexts.
3.Document Image Retrieval System with Word Recognition I
In this web site a Document Image Retrieval System (DIRS) is presented. The used technique encounters the document retrieval problem using a word matching procedure. This technique performs the word matching directly in the document images bypassing OCR and using word-images as queries. The entire system consists of the Offline and the Online procedures. In the Offline procedure, the document images are analyzed and the results are stored in a database. Three main stages, the preprocessing, the word segmentation and the feature extraction stages, constitute the offline procedure. A set of features, capable of capturing the word shape and discard detailed differences due to noise or font differences are used for the word-matching process. The Online procedure consists of four components: the creation of the query image, the preprocessing stage, the feature extraction stage, and finally, the matching procedure.
4.Content-Based Image Retrieval  - SIMPLIcity 
This content-based image search engine was developed at Stanford University between 1999 and 2000. The line of research is on-going at Penn State. Main publications include
  • Jia Li, James Z. Wang, ``Automatic linguistic indexing of pictures by a statistical modeling approach,'' IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 25, no. 9, pp. 1075-1088, 2003. (download)
  • James Z. Wang, Jia Li, Gio Wiederhold, ``SIMPLIcity: Semantics-sensitive Integrated Matching for Picture LIbraries,'' IEEE Trans. on Pattern Analysis and Machine Intelligence, vol 23, no.9, pp. 947-963, 2001. (download)
The current database has about 200,000 images from COREL CD-ROM Collection.
5.Behold | Search High Quality Flickr Images
Behold is a search engine for high-quality Flickr images. It aims to answer your queries based on what is inside the images -- at the pixel level. It offers a completely new way to search for images, using techniques of computer vision. It is different to standard image search engines, such as Flickr or Google, because those search through images using only image tags and filenames.

Behold looks for high quality images, so you don't have to sift through hundreds of poorly taken pictures to find a good one. Behold uses both aesthetic and technical quality indicators to find some of the best images available online.

Behold draws computational power from Amazon Elastic Compute Cloud (EC2) to handle large volumes of images.

6.ALIPR - Automatic Photo Tagging and Visual Image Search
The ALIPR (pronounced a-lip-er), launched officially on November 1, 2006, is a machine-assisted image tagging and searching service being developed at Penn State by Professors Jia Li and James Z. Wang. They started their work as early as in 1995 while they were both with the Stanford University. They attempted to develop computer systems to manage millions of images by the pixel content. They have developed the WIPE (TM) system, the first good-accuracy image-based parental control filter for Web images, in 1997. They have also developed the SIMPLIcity (TM) image similarity search engine, handling millions of images in real-time, in 1999 (note: this is unrelated to the later Picture Simplicity - Picasa work by Google).

7.alphaWorks : IBM Multimedia Analysis and Retrieval System
IBM Multimedia Analysis and Retrieval System (IMARS) is a powerful system that can be used to automatically index, classify, and search large collections of digital images and videos. IMARS works by applying computer-based algorithms that analyze visual features of the images and videos, and subsequently allows them to be automatically organized and searched based on their visual content. In addition to search and browse features, IMARS also:
  • Automatically identifies, and optionally removes, exact duplicates from large collections of images and videos
  • Automatically identifies near-duplicates
  • Automatically clusters images into groups of similar images based on visual content
  • Automatically classifies images and videos as belonging or not to a pre-defined set (hereafter called taxonomy) of semantic categories (such as ‘Landmark’, ‘Infant’, etc.)
  • Performs content-based retrieval to search for similar images based on one or more query images
  • Tags images to create user defined categories within the collection
  • Performs text based and metadata based searches.

8.retrievr - search by sketch / search by image
retrievr is an experimental service which lets you search and explore in a selection of Flickr images by drawing a rough sketch.
Currently the index contains many of Flickr's most interesting images. If you'd like to have your images (or the images for a specific tag) added, please let me know. A submission interface is planned!

9.Pixolu - find what you imagine
pixolu is a prototype of an image search system combining keyword search with visual similarity search and semi-automatically learned inter-image relationships. Enter a search term and pixolu searches the image indexes of Yahoo and Flickr. Compared to other image search systems pixolu retrieves more images in an initial phase. Due to a visually sorted display up to several hundred images can be easily inspected, which in most cases is sufficient to get good representations of the entire result set. The user can quickly identify images, which are good candidates for his/her desired search result.In the next step the selected candidate images are used to refine the result by filtering out visually non-similar images from an even larger result set. In addition pixolu learns the inter-image relationships from the candidate sets of different users. This helps pixolu to suggest other images that are semantically similar to the candidate images.

10.imprezzeo | find the right image
Imprezzeo is an image-based search technology company. Our technology allows users to search for images using other images as examples rather than textual search terms. Those images might be scenes, landmarks, objects, graphics, people or even personalities. Irrespective of the size of the collection, Imprezzeo Image Search helps you find the right image, fast.
By delivering a more satisfying search experience, Imprezzeo helps content providers reduce abandoned search sessions, increase usage and improve customer loyalty.
11.CoPhIR - COntent-based Photo Image Retrieval
The CoPhIR (Content-based Photo Image Retrieval) Test-Collection has been developed to make significant tests on the scalability of the SAPIR project infrastructure (SAPIR: Search In Audio Visual Content Using Peer-to-peer IR) for similarity search.
CoPhIR is now available to the research community to try and compare different indexing technologies for similarity search, with scalability being the key issue.
The organizations (universities, research labs, etc.) interested in building experimentations on it should sign the enclosed CoPhIR Access Agreement and the CoPhIR Access Registration Form, sending the original signed documents to us by mail. Please follow the instruction in the section “How to get CoPhIR Test Collection”. You will then receive Login and Password to download the required files. 

In this web-site a new set of feature descriptors is presented in a retrieval system. These descriptors have been designed with particular attention to their size and storage requirements, keeping them as small as possible without compromising their discriminating ability. These descriptors incorporate color and texture information into one histogram while keeping their sizes between 23 and 74 bytes per image. Also, in this web-site an Auto Relevance Feedback (ARF) technique is introduced which is based on the proposed descriptors. The goal of the proposed Automatic Relevance Feedback (ARF) algorithm is to optimally readjust the initial retrieval results based on user preferences. During this procedure the user selects from the first round of retrieved images one as being relevant to his/her initial retrieval expectations. Information from these selected images is used to alter the initial query image descriptor.

13.Google Similar Images
Google Similar Images is an experimental service from Google Labs that lets you find images that are similar with an image you select.
"Similar Images allows you to search for images using pictures rather than words. Click the Similar images link under an image to find other images that look like it."

img(Rummager) brings into effect a number of new as well as state of the art descriptors. The application can execute an image search based on a query image, either from XML-based index files, or directly from a folder containing image files, extracting the comparison features in real time. In addition the img(Rummager) application can execute a hybrid search of images from the application server, combining keyword information and visual similarity. Also img(Rummager) supports easy retrieval evaluation based on the normalized modified retrieval rank (NMRR)
and average precision (AP).

15.Lire Demo
The LIRE (Lucene Image REtrieval) library provides a simple way to retrieve images and photos based on their color and texture characteristics. LIRE creates a Lucene index of image features for content based image retrieval (CBIR). Three of the available image features are taken from the MPEG-7 Standard: ScalableColor, ColorLayout and EdgeHistogram a fourth one, the Auto Color Correlogram has been implemented based on recent research results. Furthermore simple methods for searching the index and result browsing are provided by LIRE. The LIRE library and the LIRE Demo application as well as all the source are available under the Gnu GPL license.

No comments: