I use Pascal VOCdevkit to calculate object detection average precision for each class, but how can I get mean average precision for the whole dataset? Should I average each average precision or should I calculate recall and precision vector for the whole dataset and use the AP equation? Thanks!
I think, the Average Precision(AP) is defined for one class. Then Mean Average Precision(MAP) should simply be the mean($\mu$) over multiple classes, say 10 or so.