The paper introduces two new regional shape descriptors, 3D shape contexts and harmonic shape contexts, for recognizing objects in noisy and cluttered scenes. These descriptors are evaluated on the task of recognizing vehicles in range scans using a database of 56 cars. The performance of these new descriptors is compared to the existing spin image descriptor, showing that the shape context-based descriptors have a higher recognition rate in noisy scenes, with 3D shape contexts outperforming others in cluttered scenes. The paper also discusses the design and implementation of these descriptors, including the use of locality-sensitive hashing to speed up search times while maintaining accuracy. The experiments demonstrate the effectiveness of the new descriptors in handling noisy and cluttered conditions, with 3D shape contexts performing particularly well in cluttered scenes.The paper introduces two new regional shape descriptors, 3D shape contexts and harmonic shape contexts, for recognizing objects in noisy and cluttered scenes. These descriptors are evaluated on the task of recognizing vehicles in range scans using a database of 56 cars. The performance of these new descriptors is compared to the existing spin image descriptor, showing that the shape context-based descriptors have a higher recognition rate in noisy scenes, with 3D shape contexts outperforming others in cluttered scenes. The paper also discusses the design and implementation of these descriptors, including the use of locality-sensitive hashing to speed up search times while maintaining accuracy. The experiments demonstrate the effectiveness of the new descriptors in handling noisy and cluttered conditions, with 3D shape contexts performing particularly well in cluttered scenes.