Nowadays, depth data has an important role in many applications. The sensors which can capture depth data became essential parts of autonomous vehicles. These sensors record a huge amount of 3D data (point clouds with x, y, and z coordinates). Furthermore, for many point cloud processing applications, it is important to calculate feature vectors that aim at describing the neighborhood of each point. Usually, a feature vector has high dimensionality, and storing it in a database is a difficult task. One of the most common operations on feature descriptors is the nearest neighbor search. However, earlier works show that nearest neighbor search with spatial index structures in high dimensions could be outperformed by sequential scan. In this work, we investigate how dimensionality reduction on 3D feature descriptors affects the descriptiveness.

Eredeti fellelhetőség: link.springer.com