SemanticMetadata.net

Content Based Image Resizing (GPL, Java): Update to v3

I’ve just uploaded a new version of the seam carving application (v2 v3). There is a performance increase and the result should be smoother now. It’s either available as Java Webstart or as compiled jar file to start the Swing GUI. Sources are included, license is GPL, Java 1.6 needed.

• Java Webstart: ImageSeams
• Download binaries & source v2 (43k)
• Download binaries & source v3 (Java Swing GUI App, 54k)
• Download Windows binary (Java Swing GUI App with Windows launcher, Java 1.6 needed, 243k)

What does this application do?

ImageSeams accepts JPG and PNG images as input and re-targets the image. This means it cuts out the not interesting parts of an image automatically. The application is based on the findings of this group described in this paper. A visual introduction of the original application is given in this video.

Changes:

• Bugfix: Bug fixed in seams backtracking, results are smoother now
• Feature: Image can now be scrolled
• Feature: About dialog showing version
• Change: Speed up in image repaint: ~30% ~50% performance gain
• Change: Icons for seam carving replaced
• Chang: Animation is now smoother

Small tutorial:

• Load an image with the leftmost button (jpg and png will work fine, try to use images not bigger than 800×800 for a start)
• Set the preferred size using the spinners (Do not forget to hit enter if you type it in, otherwise Java won’t recognize it)
• Hit the 2nd button in the tool bar and observe the effect
• Tryout the image brushing (click and drag on the image) to retain certain parts (e.g. faces) within the image
• Switch to the remove brush if you want to remove parts of the image before resizing.

Content Based Image Resizing: Webstart GUI

To allow easy tryouts I just uploaded a Java Webstart version of the ImageSeams GUI tool. You need to have Java 1.6 installed, then you just click here and the application should start right away.

• Java Webstart ImageSeams

If you just want to know what this is all about … here is a short flash demo showing the brush and resize trick.

First steps are:

• Load an image with the rightmost button (jpg and png will work fine, try to use images not bigger than 800×800 for a start)
• Set the preferred size using the spinners (Do not forget to hit enter if you type it in, otherwise Java won’t recognise it)
• Hit the 2nd button in the toolbar and observe the effect
• Tryout the image brushing (click and drag on the image) to retain certain parts (e.g. faces) within the image
• Switch to the remove brush if you want to remove parts of the image before resizing.

Content Aware Image Resizing: GPL Implementation

Just for the proof of concept I’ve implemented parts of the Content Aware Image Resizing algorithm (presented by Avidan, S. & Shamir, A.) shown here (covered also on Slashdot, Techcrunch, …) in Java (GPL). Currently image size can only be reduced.

The images on the left show results from the implementation. The source as well as the binaries can be downloaded here:

• Source & Binaries (Java 1.6 needed, ~30k): SeamCarving.tar.bz2SeamCarving.zip
• Windows Executable (Java 1.6 needed, ~200k): SeamCarving-Win-Bin.zip
• The most current implementation can be found here.

I’ve furthermore put up a GUI tool for loading and ‘carving’ images. It allows to mark areas for retaining (red brush) as well as for removal (green brush). The screenshot shows the application with area already colored. Download is here:

• Source & Binaries (Java 1.6 needed, ~65k): ImageCarving-GUI.zip

LIRe 0.5.4 Released

In Lire 0.5.4 some bugs were fixed: The scalable color descriptor (color histogram) was not compliant to the MPEG-7 standard, which is now fixed. The color only search was changed to use the color layout descriptor and a bug in the edge histogram descriptor was hunted down.

Note that you have to re-index your files: Your old index cannot be used with the new version as 2 descriptors have changed. Furthermore all binary files have been compiled with Java 6.0. So if you need a Java 1.5 version you’ll need to recompile yourself (ant build file is included) and include the swing layout class library from NetBeans.

The LireDemo GUI application has also been updated: A new function for creating image mosaics has been introduced and the indexing of digital photos is now faster than ever as only the EXIF thumbnails - if available - are used instead of the whole image.

Go to the LIRe page for download links and further information.

Caliph & Emir v0.9.25 released

Caliph & Emir 0.9.25 is mainly a maintainance release: There were some bugs in the content based image retrieval classes. Scalable Color descriptor extraction was fixed and yields now the same results as the MPEG-7 reference software. Furthermore ppm and bmp images are now supported. Note that Java 6.0 is required for the binary version and 1.5 for building it yourself.

The Scalable Color Descriptor was fixed: Now it matches the XM implementation (thanks to Fabrizio Falchi and Bastian Hoesch). EdgeHistogramImpl has also been updated according to a bug report & patch by Bastian Hoesch. Furthermore the DominantColor class was adapted based on suggestions by Janine Lachner. New feature is the support for ppm and bmp images.

Binaries, source and an installer for windows can be downloaded on sourceforge.net.

Lire 0.5.2 Released: Auto Color Correlogram

The 0.5.2 release of LIRe brings along a new descriptor, which is kind or “more advanced version of a color histogram”. The  so called color correlogram is based on the probability to find pixels of certain colors in certain neighborhoods. Leaving the theoretical part aside the color correlogram is a new way to retrieve photos with LIRe based on color and color distribution, which might be very interesting for applications heavily depending on colors. Further information on the correlogram might be found at the development Wiki.

Lire 0.5.1 released

Today I released Lire 0.5.1: Since Lire already supports Color histograms (with the MPEG-7 ScalableColor descriptor), functions for searching for colors have been integrated by by adding a searcher for color only search operations, a document builder restricted to color and a document factory for fast and efficient creation of documents describing images with one color only.

Download Lire at the sourceforge.net project page.

Lire 0.5 Released feat. Identification of Duplicate Images

The 0.5 release of lire includes one major bug fix as well as a new feature. Lire can now easily be used to identify duplicate images within an index. Sample code and documentation on the new feature can be found in the documentation wiki. The API docs are online available.

The bug, which is now fixed, was responsible for missing results in the result list in the special case of equal distance to the query image. Additionally the Lucene library packaged with Lire has been updated to version 2.0.0.

Download the new release at sourceforge.net:

• Lire-0.5.tar.bz2
• Lire-0.5.zip

Continue reading Lire 0.5 Released feat. Identification of Duplicate Images…

Caliph & Emir 0.9.23 Released

This release focuses on the visualization of repositories and search results in Emir. Alt + mouse wheel now changes size of thumbnails in the visualization and the movement within the visualization has been fixed. It further includes library updates and a new demo mode (endless loop of iterative arrangement in visualization) in Emir.

The new release can be downloaded at sourceforge:

• CaliphEmir_v0.9.23-bin.tar.bz2
• CaliphEmir_v0.9.23-bin.zip
• CaliphEmirSetup_v0.9.23_build339.exe

Lire (Lucene Image REtrieval) 0.4 Released

The LIRE library allows the creation of a Lucene Index for content based image retrieval (CBIR). Furthermore methods for searching the index are provided. In Lire 0.4 support for weighting has been added to the image searcher module. With weighting the importance of texture, colour and colour distribution for image retrieval can be specified. The new feature is described in detail in the online documentation and the API docs.

Download:

• Lire-0.4.tar.bz2 (833072 Bytes)
• Lire-0.4.zip (926042 Bytes)

Continue reading Lire (Lucene Image REtrieval) 0.4 Released…