Automatic 3D Cranial Landmark Positioning based on Surface Curvature Feature using Machine Learning

Cranial anthropometric reference points (landmarks) play an important role in craniofacial reconstruction and identification. Knowledge to detect the position of landmarks is critical. This work aims locate automatically. Landmarks positioning using Surface Curvature Feature (SCF) inspired by conventional methods finding based on morphometrical features. Each cranial landmark has a unique shape. With appropriate 3D descriptors, computer can draw associations between shapes machine learning. The challenge classification detection three-dimensional space determine model data representation. Using raw learning serious volumetric issue. uses as descriptor. It extracts local surface curvature shape into projection sequential value (depth). A method developed characteristics. Classification carried out from top-n prediction probabilities for each class, set predictions, then filtered get pinpoint accuracy. are hypothetically clustered particular area, so cluster-based filter isolate them. successfully detected landmarks, with average distance ground truth being 0.0326 normalized units. implemented increase accuracy compared truth. Thus, SCF suitable descriptor landmarks.