So there it is, I want to get a camera video stream from a server, and be able to know how many people appear on it.
I can install any OS (MAC, Windows, Linux) on my server, use any API (free or paid) and any language (C++, php etc.) to do so.
I already found some interesting things, such as OpenCV but I think it's more useful for face recognizing than to know the number of people in a room...
Any help would be appreciated :)
Use a human detector applied on each frame for basic statistic exports:
OpenCV has one already:
http://docs.opencv.org/doc/tutorials/objdetect/cascade_classifier/cascade_classifier.html
All you have to do is replace the face model with a human body (_fullbody.xml):
https://github.com/Itseez/opencv/tree/master/data/haarcascades
You can now export some statistics such as how many people per frame there are. To find how many unique individuals appeared in a sequence, you'll need a tracker. From there things get more difficult. You'll have to find something that suits your case. Tip: google for "multi-target tracking code" or similar.
You can also use openCV's latent svm detector: http://docs.opencv.org/modules/objdetect/doc/latent_svm.html
with a person model from here: https://github.com/Itseez/opencv_extra/tree/master/testdata/cv/latentsvmdetector/models_VOC2007
来源：https://stackoverflow.com/questions/18461211/know-how-many-people-are-on-a-video