imgSeek is a photo collection manager and viewer with content-based search and many other features.
The query is expressed either as a rough sketch painted by the user or as another image.
The searching algorithm makes use of multi-resolution Haar wavelet decomposition of the query and database images.
An open source server-side version of the same image similarity engine found on imgSeek is also available.
The isk-daemon database server capable of adding content-based (visual) image searching to any image related website or software through a XML-RPC or SOAP HTTP-based API.
Projects:
F.A.Q.
The first time you run imgSeek, an empty database is created, and later saved to the file "~/.imgseek/img-db.iqd". You can populate it on the "Add images" tab.
Before clicking on the "Add" button, you should choose type (or click on the "..." button) the full path of the directory you wish to add. Then click the "Add" button and wait. A progress dialog will appear so you can see how the process is going or Abort it.
This process is extremely slow for large photo collections, so please be patient. On a next version I'll try hard to make it faster.
Sometimes it is useful to filter the files you add, for example, to exclude any thumbnails dirs and files, you could enter "tn" on the string restriction and "50" for the size.
See more detailed instructions.
The query is expressed either as a rough sketch painted by the user or as another image.
The searching algorithm makes use of multi-resolution Haar wavelet decomposition of the query and database images.
An open source server-side version of the same image similarity engine found on imgSeek is also available.
The isk-daemon database server capable of adding content-based (visual) image searching to any image related website or software through a XML-RPC or SOAP HTTP-based API.
Projects:
- Desktop version, a photo collection manager and viewer with content-based search and many other features;
- Server version: for integrating a content-based image database into your own software or image-related web site.
- When starting imgSeek, all I get is: "Warning: Unable to load the C++ extension "imgdb.so" module. Make sure ...
- I get an error when exiting: Fatal Python error: PyEval_RestoreThread: NULL tstate
- The metadata editor ignores or refuses to save comments with extended characters on my locale.
- I have an nvidia card and I get the following error: "ImportError: libnvidia-tls.so.1: cannot handle TLS data"
- I get an error when running on a Mandrake system: "ImportError: No module named libsip"
- How do I run a similarity match on only the images in a some directory (and its subdirectories)?
- How do I do a similarity search which returns only matches that are NOT in the same image set?
- Why Qt and PyQt ?
- How do I install PyQt ?
- Why ImageMagick or QT development files are required ?
- I'm trying to compile PyQT, but it fails with "sipqtQDesktopWidget.cpp:4079: no method QDateTimeEdit::layoutEditors"
- I couldn't get imgSeek working because of PyQt related errors, or it simply segfaults...
- Running imgSeek inside gdb
- QT libraries ? ImageMagick ? Development files ? What do I need to run imgSeek ?
- How do I install imgSeek if I don't have root privileges ?
- Where are database files located, what is their size and how do they scale?
- Why metadata embedded on image files don't get updated when I change them inside imgSeek ?
- I'm not able to extract IPTC metadata and use the Transform tool dialog.
- I would like to see the feature XYZ in imgSeek !
- How do I create an RPM package for imgSeek ?
- How can I help imgSeek development ?
- What about running it on Windows(R) systems ?
- For what platform is imgSeek being designed?
- Do we expect imgSeek to be used as a server-side image collection manager ?
- Install fails with a message similar to "cc1plus: unrecognized option `-fI/usr/include'"
Before clicking on the "Add" button, you should choose type (or click on the "..." button) the full path of the directory you wish to add. Then click the "Add" button and wait. A progress dialog will appear so you can see how the process is going or Abort it.
This process is extremely slow for large photo collections, so please be patient. On a next version I'll try hard to make it faster.
Sometimes it is useful to filter the files you add, for example, to exclude any thumbnails dirs and files, you could enter "tn" on the string restriction and "50" for the size.
See more detailed instructions.
If you liked this article, subscribe to the feed by clicking the image below to keep informed about new contents of the blog:
0 commenti:
Post a Comment