Table 19.
Literature Analysis: 3D Place Recognition (3DPR) Methods.
| Model | Environment | Scenario | Sensors | 3D Place Recognition |
|---|---|---|---|---|
| RGBD-Net [78] | Indoor | Depth-specific features learning for scene recognition scenario | Camera | RGB-D:3D Depth Feature based |
| Event-VPR [79] | Outdoor | Event-based visual place recognition scenario in changing environment | Camera | Event-based |
| Pointnetvlad [80] | Outdoor | Point-cloud-based retrieval scenario for place recognition | LiDAR | Point Cloud based |
| ISR-Net [81] | Indoor | Indoor scene recognition scenario with 3D scene representations (point clouds or voxels) | LiDAR | Point Cloud based |
| PCPR-Net [82] | Outdoor | Point-cloud-based place recognition scenario using hierarchical features extraction with CNN | LiDAR | Point Cloud based |
| Lpd-net [83] | Outdoor | Large scale place recognition scenario with feature extraction using global descriptors | LiDAR | Point Cloud based |
| OREOS [84] | Outdoor | Oriented recognition scenario to retrieve nearby place candidates | LiDAR | Point Cloud based |
| SDM-Net [85] | Outdoor | Place recognition scenario from a scene’s structure with semi-dense point clouds | LiDAR | 3D-voxel grid |
| SDes-Net [86] | Outdoor | 3D segment based on learned descriptors for place recognition scenario | LiDAR. | Point Cloud based |
| MinkLoc3D [87] | Outdoor | Place recognition scenario with discriminative 3D point cloud descriptor. | LiDAR. | Sparse voxelized point-cloud-based |
| CLFD-Net [88] | Outdoor | Fused global feature generation scenario for place recognition scenario | Camera, LiDAR | Image and Point Cloud based Fusion |
| PIC-Net [89] | Outdoor | Fusion based Place recognition scenario based on image and point clouds | Camera, LiDAR | Image and Point Cloud based Fusion |