Local Feature Extraction Network for Point Cloud Analysis

Geometric feature extraction of 3D point clouds plays an important role in many 3D computer vision applications such as region labeling, 3D reconstruction, object segmentation, and recognition. However, hand-designed features on point clouds lack semantic information, so cannot meet these requirements. In this paper, we propose local feature extraction network (LFE-Net) which focus on extracting local feature for point clouds analysis. Such geometric features learning from a relation of local points can be used in a variety of shape analysis problems such as classification, part segmentation, and point matching. LFE-Net consists of local geometric relation (LGR) module which aims to learn a high-dimensional local feature to express the relation between points and their neighbors. Benefiting from the additional singular values of local points and hierarchical neural networks, the learned local features are robust to permutation and rigid transformation so that they can be transformed into 3D descriptors. Moreover, we embed prior spatial information of the local points into the sub-features for combining features from multiple levels. LFE-Net achieves state-of-the-art performances on standard benchmarks including ModelNet40, ShapeNetPart.

Point cloud clustering and outlier detection based on spatial neighbor connected region labeling

Clustering analysis is one of the most important techniques in point cloud processing, such as registration, segmentation, and outlier detection. However, most of the existing clustering algorithms exhibit a low computational efficiency with the high demand for computational resources, especially for large data processing. Sometimes, clusters and outliers are inseparable, especially for those point clouds with outliers. Most of the cluster-based algorithms can well identify cluster outliers but sparse outliers. We develop a novel clustering method, called spatial neighborhood connected region labeling. The method defines spatial connectivity criterion, finds points connections based on the connectivity criterion among the k-nearest neighborhood region and classifies connected points to the same cluster. Our method can accurately and quickly classify datasets using only one parameter k. Comparing with K-means, hierarchical clustering and density-based spatial clustering of applications with noise methods, our method provides better accuracy using less computational time for data clustering. For applications in the outlier detection of the point cloud, our method can identify not only cluster outliers, but also sparse outliers. More accurate detection results are achieved compared to the state-of-art outlier detection methods, such as local outlier factor and density-based spatial clustering of applications with noise.

Adding Context Information to Video Analysis for Surveillance Applications

Smart surveillance systems become more meaningful if they both grow in reliability and robustness, while simultaneously offering a higher semantic level of understanding. To achieve a higher level of semantic scene understanding, the objects and their actions have to be interpreted in the given context, so that the extraction of contextual information is required. This chapter explores several techniques for extracting the contextual information such as spatial, motion, depth and co-occurrence, depending on applications. Afterwards, the chapter provides specific case studies to evaluate the usefulness of context information, based on: (1) region labeling of the surroundings of objects, (2) motion analysis of the water for moving ships, (3) traffic sign recognition for safety event evaluation and (4) the use of depth signals for obstacle detection. The chapter shows that the previous cases can be solved in an improved way with respect to robustness and semantic understanding. Case studies indicate up to 6.8% improvement of reliable correct object understanding and the novel possibility of labeling scene events as safe/unsafe depending on the object behavior and the detected surrounding context. In this chapter, it is shown that using contextual information improves automated video surveillance analysis, as it not only improves the reliability of moving object detection, but also enables scene understanding that is far beyond object understanding.

Adding Context Information to Video Analysis for Surveillance Applications

Smart surveillance systems become more meaningful if they both grow in reliability and robustness, while simultaneously offering a higher semantic level of understanding. To achieve a higher level of semantic scene understanding, the objects and their actions have to be interpreted in the given context, so that the extraction of contextual information is required. This chapter explores several techniques for extracting the contextual information such as spatial, motion, depth and co-occurrence, depending on applications. Afterwards, the chapter provides specific case studies to evaluate the usefulness of context information, based on: (1) region labeling of the surroundings of objects, (2) motion analysis of the water for moving ships, (3) traffic sign recognition for safety event evaluation and (4) the use of depth signals for obstacle detection. The chapter shows that the previous cases can be solved in an improved way with respect to robustness and semantic understanding. Case studies indicate up to 6.8% improvement of reliable correct object understanding and the novel possibility of labeling scene events as safe/unsafe depending on the object behavior and the detected surrounding context. In this chapter, it is shown that using contextual information improves automated video surveillance analysis, as it not only improves the reliability of moving object detection, but also enables scene understanding that is far beyond object understanding.