If I am not wrong, IDOL is working to release improvements in new capabilities in the next six months. There is room for improvement in some very important capabilities in visual analytics. They have been focusing on improving the face recognition algorithm. The accuracy of object detection could be improved as well and I know they are working on that at the moment. I would like to see some machine learning capabilities added to the next release.
