Though seemly a simple task, composing good photographs are often tedious and painful. Insight automatically composes the photographs you have already shot with a pre-defined aspect ratio using a set of computer vision algorithms.
The core of Insight is these algorithms. They try to find the most visually pleasing region inside photographs.<p>We will be releasing a public API very soon. In the mean while please do play around with demo. Your feedback is extremely important for me! Specifically I would like to know your thoughts on:<p>* Is there a need for such services.
* The performance of the algorithm The algorithm itself is undergoing active revisioning. Both positive and negative feedback about the suggestions give by the algorithm will be very much appreciated.
* Please drop me a line in case you are interested to beta-test the API.
This is awesome. I had to crop user generated photos for thumbnails in a photo gallery and a lot of times it would miss human faces. Does this guarantee it won't happen?
Yes, in most cases the algorithm will ensure that the face is included. However, there are a few instances ( for example, if the face is blurred, or is in a weird pose ) when it will miss it.
It would be great if a user could mark some important points in the image. The algorithm could then ensure that these are included prominently in the resulting crop.
