This work deals with hip joint cartilage segmentation, which is an important task in joint diseases diagnosis. Since the manual segmentation commonly used nowadays, is a tedious and lengthy task, this work brings a new idea into its automation and simplification of the medical expert work. The proposed method is an interactive algorithm for hip joint cartilage segmentation, and hence it requires some user interaction, which is common and usually desired in medical image processing. The first step of the method is a selection of several points in several slices lying inside the cartilage. The purpose is a rough delimitation of the cartilage area. This delimitation is based on the assumption that the femoral head has approximately circular shape. The cartilage is then segmented using the combination of Graph Cut segmentation and K-means clustering techniques. The algorithm was tested on 3D isotropic True-FISP volumes of patients with femoroacetabular impingement and the results were evaluated by commonly used Dice Similarity Coefficient (0.62). The results of the segmentation will be prospectively used for the segmentation of quantitative MR maps (T1 and T2) in femoro-acetabular impingement (FAI) investigation.