Professional's Domain

Researchers from Penn State university have released ACQUINE (Aesthetic Quality Inference Engine), an online photo-rating search engine. According to the developers of the system, ACQUINE is the first publicly available tool for automatically determining the aesthetic value of an image. Associate professor James Z. Wang of information sciences and technology is one of the principal researchers on ACQUINE, said this tool is a significant first step in recognizing human emotional reaction to visual stimulus.

The ACQUINE has been launched ever since 2005 and come to a public release in April 2009. The system uses a series visual aspects of an image to determine the aesthetic rating, including color saturation, color distribution and photo composition. On the ACQUINE web site (http://acquine.alipr.com), users can upload any image to the system and in a few seconds a rating from zero to 100 is given automatically.

The system has been trained to simulate the human examination process of images using machine learning techniques. By studying thousands of previously-rated photographs from online photo-sharing Web sites such as photo.net, the system is able to behave quite in the human way: it has achived 80 percent consistency between the human and computer ratings, and will be even better in the near future.

New extensions are also being developed such as integrating the rating system into digital camaers to help photographers to take best pictures. The new technology introduced by ACQUINE may significantly change the way we share and search images now.