Added more comments about how to pack multiple detectors into a single

object_detector.
This commit is contained in:
Davis King 2014-03-07 19:45:18 -05:00
parent b0795c962f
commit 5000b9f0ac
1 changed files with 9 additions and 1 deletions

View File

@ -239,7 +239,15 @@ int main(int argc, char** argv)
// into a single object_detector must have been trained with the same settings for
// the sliding window size and the scanner padding option (see the scan_fhog_pyramid
// documentation for a discussion of padding). This is because they all share the
// same scanner object inside the object_detector.
// same scanner object inside the object_detector.
//
// To pack multiple detectors into a single object_detector you use code like this:
std::vector<object_detector<image_scanner_type> > my_detectors;
// Add your component detectors into my_detectors.
my_detectors.push_back(detector);
// Then just construct an object_detector from the set of detectors.
object_detector<image_scanner_type> this_object_has_multiple_detectors(my_detectors);
//
//
// Finally, you can add a nuclear norm regularizer to the SVM trainer. Doing has
// two benefits. First, it can cause the learned HOG detector to be composed of