IBM Multimedia Analysis and Retrieval System

1. What is the IBM Multimedia Analysis and Retrieval System?

IBM Multimedia Analysis and Retrieval System (IMARS) is an automated tool being developed by IBM Research for content-based indexing, classification and search for digital image and videos.

Back to top

2. What are the advantages of visual search over the common, text or tagged based tools?

IMARS enables the user to index, browse and search a collection of videos or images based on their visual content. This capability is much more powerful than the traditional text/tag based tool (which IMARS uses as a baseline), since it relies on the actual information contained in an image or video, rather than on something that some people labeled. For example, when looking for images containing orange, visual cues such as its color provide a mean to get far more meaningful results than simply the name of the picture itself, which often carries no information about its content (i.e. PICT_001.jpg)

Back to top

3. Does IBM Multimedia Analysis and Retrieval System work on videos?

Yes. The IMARS natively supports digital images of most prevalent image formats and video in MPEG1 and MPEG2 formats. Other video formats can be supported using the FFmpeg video decoder. Instructions for using FFmpeg within IMARS system are provided in the user guide.

Back to top

3. Why is there a limitation on the number of images and videos that the system can process in a single batch?

This is not a software limitation, but merely a trial license limitation for this release. The tool can, however, be run on an unlimited number of image and video collections (limited only by the available disk storage), and the collection size limitation should be generous enough to allow most non-commercial user applications.

Back to top

4. If I check the Search subdirectory box in the IBM Multimedia Analysis and Retrieval System collector, what is the maximum depth of subdirectories that will be visited?

Maximum depth for crawling subdirectories is 12.

Back to top

4. What does IBM Multimedia Analysis and Retrieval System consist of?

IMARS is comprised of the IMARS extraction tool and the IMARS search tool. The IMARS extraction tool is run against a collection of images and videos, and it generates indices based on computer analysis of the collection and organizes the results for subsequent searching, browsing and navigation using the IMARS search tool.

The system consists of two components: Indexing and Search. When you click on Imars.exe in the <DIR>/bin directory, the indexing and ingestion tool is displayed. After the indexing and ingestion are complete, the search tool can be accessed at http://localhost/imars/ through any browser.

You cannot use the search tool without running the indexing and ingestion tool at least once.

Back to top

5. Why does the system work only on the Windows® platform?

The system's indexing tool works only on Windows. However, the system's search tool can be accessed from any Web browser on any platform (Windows, Apple, Linux®, etc.).

Back to top

5. Why is there a limitation on the number of images and videos that the system can process in a single batch?

The trial version of IMARS limits the number of images and videos that can be processed, but should be generous enough for most non-commercial testing of the system. The full version of IMARS is limited only by the available disk space.

Back to top

6. If I check the Search subdirectory box in the IMARS extraction tool, what is the maximum depth of subdirectories that will be visited?

Maximum depth for crawling subdirectories is 12.

Back to top

7. Can I append another image collection to the one ingested in IBM Multimedia Analysis and Retrieval System?

Not at this time. This release allows you to run the system on separate directories only. You can put two different photo collections in the same folder and run the system on them (make sure to check the include subfolders option).

Back to top

7. Why does IBM Multimedia Analysis and Retrieval System work only on the Windows XP platform?

The IMARS extraction tool of the IBM Multimedia Analysis and Retrieval system has been extensively tested on Windows XP platform. It will run on any 32-bit Windows Operating System, but we recommend Windows XP. However, the IBM Multimedia Analysis and Retrieval System Search Tool can be accessed from any Web browser on any platform, including Windows, Apple, and Linux.

Back to top

8. Do I need to install a Web server? Why?

Yes. In order to use the IBM Multimedia Analysis and Retrieval System Search Tool, you do need to install the Apache http server, please refer to the “Apache HTTP Server Setup” section of the IMARS User Guide (PDF).

Back to top

9. Can I append another image collection to the one ingested in IBM Multimedia Analysis and Retrieval System?

Not at this time. This release allows you to run IBM Multimedia Analysis and Retrieval System on separate directories only. You can put two different photo collections in the same folder and run IBM Multimedia Analysis and Retrieval System on them (make sure to check the include subfolders option).

Back to top

10. My instance of the IBM Multimedia Analysis and Retrieval System exits without warning; what should I do?

Please download the latest version of IBM Multimedia Analysis and Retrieval System from IBM alphaWorks website. If you cannot find answers in this guide, please contact us.

Back to top

10. Why isn't an index.html file generated in the top-level output directory, pointing to <IMARSDIR>/docs/html/main/0/index.html?

An index.html file is generated in the <IMARSDIR>/docs/html/ directory that points to <IMARSDIR>/docs/html/main/0/index.html.

Back to top

11. Can I append another image collection to the one ingested in IBM Multimedia Analysis and Retrieval System?

No. This release allows you to run IBM Multimedia Analysis and Retrieval System on separate directories only. You can put two different photo collections in the same folder and run IBM Multimedia Analysis and Retrieval System on them (make sure to check the include subfolders option).

Back to top

11. Why do I need to download and unzip classifiers.zip file?

The IMARS extraction tool uses semantic classifiers located in <DIR>/bin/classifiers to evaluate against the user’s multimedia collection. If these files are missing, the user will not be able to explore the full capability of the IBM Multimedia Analysis and Retrieval system which relies on automatic semantic classification.

Back to top

12. Can I get a different skin for IBM Multimedia Analysis and Retrieval System Web search interface?

Yes: In the Templates Tab of IMARS.exe, you can find seven different colors to choose from.

Back to top

12. What are the concepts present in the taxonomy, and therefore recognized by IMARS’s classifiers?

Fully-automatic approaches based on statistical modeling of low-level visual content features have been applied for detecting the semantic concepts in our pre-defined taxonomy such as sky, party, for example.

Statistical modeling requires large amounts of annotated examples for training. Since this scenario is not applicable to unlabeled content in the user collection, we adopt an approach based on automatic semantic tagging. We re-use our existing semantic models, which were trained on various multimedia content and automatically associate confidence scores to unseen data with the cross-domain concept models. To enable cross-domain usability, we chose the general semantic models from our taxonomy, preserving consistent definitions of concepts across different multimedia and video domains such as albums, blogs, and web video.

Taxonomy structure organizes a set of general semantic category models, where each category corresponds to a node in the hierarchy that is connected to parent and children nodes. Parent-child links can correspond to relationship based on containment, for example, where each child node is a specialized variation of its parent. Our taxonomy structure allows for a different aspect for decomposing or hierarchically relating the categories corresponding to who, where, how description of the visual content, thus organizing category models according to Activity, People, Setting and Type.

Back to top

13. Can I define my own category for the classifiers?

Currently IMARS classifiers work only on the categories pre-defined in our taxonomy, and do not allow the user to build classifiers for new categories. The user can choose, among the offered ones, the classifier referring to the concept which is most similar to what she is looking for. Furthermore, though tagging, the user can create and define its own clusters of images.

Back to top

13. My instance of IBM Multimedia Analysis and Retrieval System exits without warning; what should I do?

Please download the latest version of IBM Multimedia Analysis and Retrieval System.

Back to top

14. I have installed IMARS system with default CATALOGROOT location C:\WWW. Now I want my http://localhost/imars to point to a different location. How can I change this?

By default, the CATALOGROOT variable is C:\WWW. Refer to IMARS extraction tool Interface, Manual setup for more information on how to change the CATALOGROOT variable in the IMARS extraction tool setup, and refer to the Apache HTTP Web Server setup to learn how to change CATALOGROOT in the http.conf file to the new value.

Back to top

14. Why do I get the following security warning when I browse IBM Multimedia Analysis and Retrieval System Web pages: To help protect your security, Internet Explorer has restricted this file from showing active content that could access your computer....?

If you have Windows XP Service Pack 2 installed, you might get this security warning when you try to browse any of the generated Web pages. There are no scripts or Active X controls to cause this message, and there is no consequence if you simply ignore it.

Back to top

15. I played with manual setup in preflast.txt and now the tool does not run correctly. How do I reset the tool to load default variables?

If you ran IMARS.exe, it will save the current state in <DIR>\bin\preflast.txt and it will use it for subsequent runs. Do the following:

• EXIT Imars.exe
• Erase <DIR>\bin\prefdlast.txt so that the default options in <DIR>\bin\prefdflt.txt can take effect next time Imars.exe is run.

Back to top

16. What directory should I install IMARS in?

We recommend a short directory name like c:\imars. The directory tree in which the classifiers are installed under your IMARS directory can be quite deep, and in extreme cases, you may experience errors if you have directory paths which are hundreds of characters long.

Back to top