This paper presents a novel method for recognizing free-form 3D objects in point clouds, which differs from traditional approaches that rely on local point descriptors. The proposed method creates a global model description using oriented point pair features and matches this model locally through a fast voting scheme. The global model description consists of all model point pair features, representing a mapping from the feature space to the model, where similar features are grouped together. This representation allows for the use of sparser object and scene point clouds, leading to faster performance. Recognition is performed using an efficient voting scheme on a reduced two-dimensional search space. The method is evaluated on both synthetic and real datasets, demonstrating high recognition rates and efficiency in the presence of noise, clutter, and partial occlusions. Compared to state-of-the-art approaches, the method achieves better recognition rates with significantly reduced processing times. The paper also discusses related work, including global and local methods for 3D object recognition, and provides a detailed description of the model description and voting scheme. The results show that the proposed method outperforms existing techniques in terms of both recognition accuracy and speed.This paper presents a novel method for recognizing free-form 3D objects in point clouds, which differs from traditional approaches that rely on local point descriptors. The proposed method creates a global model description using oriented point pair features and matches this model locally through a fast voting scheme. The global model description consists of all model point pair features, representing a mapping from the feature space to the model, where similar features are grouped together. This representation allows for the use of sparser object and scene point clouds, leading to faster performance. Recognition is performed using an efficient voting scheme on a reduced two-dimensional search space. The method is evaluated on both synthetic and real datasets, demonstrating high recognition rates and efficiency in the presence of noise, clutter, and partial occlusions. Compared to state-of-the-art approaches, the method achieves better recognition rates with significantly reduced processing times. The paper also discusses related work, including global and local methods for 3D object recognition, and provides a detailed description of the model description and voting scheme. The results show that the proposed method outperforms existing techniques in terms of both recognition accuracy and speed.