Color-Based Segmentatation Evaluation

This web-service performs Color-Based Segmentatation Evaluation as used in the GREC 2007 Arc Segmentation Contest and described in the paper Pixel-Accurate Representation and Evaluation of Page Segmentation in Document Images (ICPR 2006). If you have any questions, please contact Faisal Shafait.

Form Interface

We recommend you upload png files. (Other formats work, but may be too large or lossy.) Please avoid spaces or other special characters in your filenames.

File1 = Ground Truth (max. 2MB):

File2 = Segmented Image (max. 2MB):

Programmatic Interface

To submit your image programmatically, you can simply POST to this URL; the images should be parameters named "imagefile1" and "imagefile2".

From the command line, you can do this using: 

curl -D header.out -F 'imagefile1=@input1.png;type=image/png' -F 'imagefile2=@input2.png;type=image/png' http://demo.iupr.org/segeval/ > output.txt

You can also do this easily using the HTTP implementation in your favorite programming language (C#, Python, Java, Perl, etc.).