Hello everybody! I have to implement a virtual changing room. I have a camera and a green screen. Software has to be able to detect the person in front of the green screen. Then, it has to localize characteristic points of human body. The user chooses clothing, and he can virtually try it. He has to be able to move and rotate. Segmentation of the image is done and it works ok. I have a human silhouette. Can anyone suggest (I'm a newbie in the field of image processing) a proper way to automatically localize the points like hips and shoulders?