An Automatic Evaluation Method of Image Mattes

T. Kobayashi, H. Tadaaki, and N. Otsu (Japan)


Image Matting, Automatic Evaluation, Mutual Informa tion, Correlation Coefficient


In image matting, several algorithms that produce high quality results have recently been proposed; however, cer tain parameters have to be manually determined in order to obtain favorable results, which requires significant user ef fort. In this paper, we propose a method of automatically evaluating alpha mattes based on either of the two crite ria: mutual information or correlation coefficient between the matte and the image gradient fields, and selecting opti mal parameter values. This drastically reduces the user ef fort required for tuning parameters in image matting. The experimental results using several matting algorithms for various images show that the automatically selected alpha mattes are similar (or the same) to the manually selected optimal ones.

Important Links:

Go Back