Matches in SemOpenAlex for { <https://semopenalex.org/work/W4387682316> ?p ?o ?g. }
Showing items 1 to 83 of
83
with 100 items per page.
- W4387682316 endingPage "1" @default.
- W4387682316 startingPage "1" @default.
- W4387682316 abstract "LiDAR-based 3D object detection constitutes a fundamental component of autonomous driving technology. In this research, we propose a novel approach called PillarNet++ to tackle the challenges associated with fine-grained information loss during point cloud encoding and the inadequate interaction or incomplete fusion of feature maps across different scales in subsequent feature extraction stages, resulting in a decrease in partial occlusion and long-distance 3D object detection accuracy, leading to false and missed detections. The PillarNet++ method primarily comprises two modules: the Multi-Attention-Pillar-Encoding (MAPE) module and the Pseudo-image-Split-Multi-Branch-Feature-Pyramid-Networks (PSMB-FPN) module. The MAPE module enhances the information extraction capability in non-empty pillars by integrating max pooling and average pooling, by fusion of the point-wise, channel-wise, and Pillar-wise attention, the MAPE module can adaptively focus on the important information and suppress the secondary point clouds. In addition, the stacked MAPE modules can refine pillars and extract finer features. On the other hand, the PSMB-FPN module splits the pseudo-image along the channel dimension and subsequently performs Multi-Branch-Feature-Pyramid-Networks (MB-FPN) feature extraction and fusion on each channel, facilitating the interaction of multi-scale and multi-level feature maps and improving prediction accuracy. Experimental results on the KITTI 3D object detection benchmark show that the PillarNet++ method has the best performance among single-stage object detection algorithms, and even exceeds most two-stage methods." @default.
- W4387682316 created "2023-10-17" @default.
- W4387682316 creator A5036124105 @default.
- W4387682316 creator A5061550482 @default.
- W4387682316 creator A5074613177 @default.
- W4387682316 date "2023-01-01" @default.
- W4387682316 modified "2023-10-17" @default.
- W4387682316 title "PillarNet++: Pillar-based 3D object detection with multi-attention" @default.
- W4387682316 doi "https://doi.org/10.1109/jsen.2023.3323368" @default.
- W4387682316 hasPublicationYear "2023" @default.
- W4387682316 type Work @default.
- W4387682316 citedByCount "0" @default.
- W4387682316 crossrefType "journal-article" @default.
- W4387682316 hasAuthorship W4387682316A5036124105 @default.
- W4387682316 hasAuthorship W4387682316A5061550482 @default.
- W4387682316 hasAuthorship W4387682316A5074613177 @default.
- W4387682316 hasConcept C120665830 @default.
- W4387682316 hasConcept C121332964 @default.
- W4387682316 hasConcept C124101348 @default.
- W4387682316 hasConcept C127162648 @default.
- W4387682316 hasConcept C131979681 @default.
- W4387682316 hasConcept C13280743 @default.
- W4387682316 hasConcept C138885662 @default.
- W4387682316 hasConcept C142575187 @default.
- W4387682316 hasConcept C153180895 @default.
- W4387682316 hasConcept C154945302 @default.
- W4387682316 hasConcept C185798385 @default.
- W4387682316 hasConcept C192209626 @default.
- W4387682316 hasConcept C205649164 @default.
- W4387682316 hasConcept C2524010 @default.
- W4387682316 hasConcept C2776151529 @default.
- W4387682316 hasConcept C2776401178 @default.
- W4387682316 hasConcept C2781238097 @default.
- W4387682316 hasConcept C31258907 @default.
- W4387682316 hasConcept C31972630 @default.
- W4387682316 hasConcept C33923547 @default.
- W4387682316 hasConcept C41008148 @default.
- W4387682316 hasConcept C41895202 @default.
- W4387682316 hasConcept C52622490 @default.
- W4387682316 hasConcept C70437156 @default.
- W4387682316 hasConceptScore W4387682316C120665830 @default.
- W4387682316 hasConceptScore W4387682316C121332964 @default.
- W4387682316 hasConceptScore W4387682316C124101348 @default.
- W4387682316 hasConceptScore W4387682316C127162648 @default.
- W4387682316 hasConceptScore W4387682316C131979681 @default.
- W4387682316 hasConceptScore W4387682316C13280743 @default.
- W4387682316 hasConceptScore W4387682316C138885662 @default.
- W4387682316 hasConceptScore W4387682316C142575187 @default.
- W4387682316 hasConceptScore W4387682316C153180895 @default.
- W4387682316 hasConceptScore W4387682316C154945302 @default.
- W4387682316 hasConceptScore W4387682316C185798385 @default.
- W4387682316 hasConceptScore W4387682316C192209626 @default.
- W4387682316 hasConceptScore W4387682316C205649164 @default.
- W4387682316 hasConceptScore W4387682316C2524010 @default.
- W4387682316 hasConceptScore W4387682316C2776151529 @default.
- W4387682316 hasConceptScore W4387682316C2776401178 @default.
- W4387682316 hasConceptScore W4387682316C2781238097 @default.
- W4387682316 hasConceptScore W4387682316C31258907 @default.
- W4387682316 hasConceptScore W4387682316C31972630 @default.
- W4387682316 hasConceptScore W4387682316C33923547 @default.
- W4387682316 hasConceptScore W4387682316C41008148 @default.
- W4387682316 hasConceptScore W4387682316C41895202 @default.
- W4387682316 hasConceptScore W4387682316C52622490 @default.
- W4387682316 hasConceptScore W4387682316C70437156 @default.
- W4387682316 hasFunder F4320335765 @default.
- W4387682316 hasLocation W43876823161 @default.
- W4387682316 hasOpenAccess W4387682316 @default.
- W4387682316 hasPrimaryLocation W43876823161 @default.
- W4387682316 hasRelatedWork W147410782 @default.
- W4387682316 hasRelatedWork W2022849497 @default.
- W4387682316 hasRelatedWork W2407190427 @default.
- W4387682316 hasRelatedWork W2626256601 @default.
- W4387682316 hasRelatedWork W2900413183 @default.
- W4387682316 hasRelatedWork W2919210741 @default.
- W4387682316 hasRelatedWork W2953234277 @default.
- W4387682316 hasRelatedWork W3002446410 @default.
- W4387682316 hasRelatedWork W3081299480 @default.
- W4387682316 hasRelatedWork W4322096758 @default.
- W4387682316 isParatext "false" @default.
- W4387682316 isRetracted "false" @default.
- W4387682316 workType "article" @default.