Global–local articulation pattern-based pedestrian detection using 3D Lidar data

Abstract

Highly variable human poses and pedestrian occlusion make light detection and ranging (Lidar)-based pedestrian detection challenging. This letter proposes a novel framework to address these issues. Other than dividing humans into arbitrary number of parts and using the same features for all part detectors, we represent humans with globallocal articulated parts and formulate new features relying on each part's own character. Articulated parts are effective because they each usually maintain a relatively consistent shape across a broader range of body poses. In addition, to extract visible human segments from cluttered surroundings with the presence of pedestrian occlusion, both 3D information and 2D spatial information are used in a coarse-to-fine manner, making the interaction of human part and its neighbouring objects better analysed. The algorithm is evaluated over a busy street dataset and is shown to be competitive with the state-ofthe-art Lidar-based algorithms. Remarkably, even at long distances, up to 20 m, it can handle pedestrian occlusion efficiently and effectively. ARTICLE HISTORY

Learning Deformable Network for 3D Object Detection on Point Clouds

“…( 16) else (17) if the last point then (18) return nonzero voxel points in each voxel, and grid coordinates (19) else (20) continue to next point. (21) end if (22) end if (23) end if ( 24) else (25) if nz_count > K then (26) return nonzero voxel, points in each voxel, and grid coordinates. ( 27) else (28) add new voxel in hash table.…”

“…ose methods convert the irregular point clouds to regular 3D grids and apply 3D CNN detectors to realize detection task; because of the use of three-dimensional convolution, the calculation of those methods is very large. ere are also some methods [22] which project point cloud data to the perspective of aerial view and carry out 2D target detection on the image after projection. However, this kind of method loses a lot of spatial detail data, so the detection results are very limited.…”

Spatially sensitive statistical shape analysis for pedestrian recognition from LIDAR data

“…Their application on the RGB-D version of the KITTI dataset leads to state-of-the-art recognition accuracy. Du et al [6] used local-global articulated human parts and defined part-specific features. This method relies on heuristic approaches to identify upper human body and legs.…”