Matches in SemOpenAlex for { <https://semopenalex.org/work/W4386821167> ?p ?o ?g. }
Showing items 1 to 76 of
76
with 100 items per page.
- W4386821167 abstract "In 3D object detection, network prediction accuracy is greatly affected by point cloud's feature richness. However, the feature richness depends on fine-grained features extracted by the network. Currently some methods use voxel encoding approach continuously down-scaled by 3D convolution to improve the detection efficiency, but lose too many fine-grained features. Some methods directly inputting the original point cloud into the Multi-layer Perceptron (MLP) for feature extraction, which can retain more fine-grained features, but greatly reduce the detection efficiency. This work combines voxel features and point features to obtain a fused 3D map. We use an attention mechanism module that combines semantic features with spatial features to progress the former 3D feature map, which is used to constitute a richer 3D feature structure to reduce the loss of Z-axis features. Since the object geometry structure information is important for the detection task, we design a geometry-oriented auxiliary network that is jointly optimized by supervising two tasks in the training phase to guide the backbone network to understand the target structure features and discard them in the inference phase. The experiments show that our proposed detection method outperforms some previous methods in KITTI 3D/BEV detection." @default.
- W4386821167 created "2023-09-19" @default.
- W4386821167 creator A5004605359 @default.
- W4386821167 creator A5005467310 @default.
- W4386821167 creator A5009698718 @default.
- W4386821167 creator A5034946106 @default.
- W4386821167 date "2023-07-24" @default.
- W4386821167 modified "2023-09-26" @default.
- W4386821167 title "Point-Voxel Fusion for 3D Object Detection" @default.
- W4386821167 cites W2115579991 @default.
- W4386821167 cites W2798965597 @default.
- W4386821167 cites W2897529137 @default.
- W4386821167 cites W2963351448 @default.
- W4386821167 cites W2963727135 @default.
- W4386821167 cites W2968296999 @default.
- W4386821167 cites W2981949127 @default.
- W4386821167 cites W3034239557 @default.
- W4386821167 cites W3034314779 @default.
- W4386821167 cites W3035346742 @default.
- W4386821167 cites W3106250896 @default.
- W4386821167 cites W3109675406 @default.
- W4386821167 cites W3117804044 @default.
- W4386821167 cites W3166089996 @default.
- W4386821167 doi "https://doi.org/10.23919/ccc58697.2023.10240949" @default.
- W4386821167 hasPublicationYear "2023" @default.
- W4386821167 type Work @default.
- W4386821167 citedByCount "0" @default.
- W4386821167 crossrefType "proceedings-article" @default.
- W4386821167 hasAuthorship W4386821167A5004605359 @default.
- W4386821167 hasAuthorship W4386821167A5005467310 @default.
- W4386821167 hasAuthorship W4386821167A5009698718 @default.
- W4386821167 hasAuthorship W4386821167A5034946106 @default.
- W4386821167 hasConcept C131979681 @default.
- W4386821167 hasConcept C138885662 @default.
- W4386821167 hasConcept C153180895 @default.
- W4386821167 hasConcept C154945302 @default.
- W4386821167 hasConcept C2776151529 @default.
- W4386821167 hasConcept C2776214188 @default.
- W4386821167 hasConcept C2776401178 @default.
- W4386821167 hasConcept C31972630 @default.
- W4386821167 hasConcept C41008148 @default.
- W4386821167 hasConcept C41895202 @default.
- W4386821167 hasConcept C45347329 @default.
- W4386821167 hasConcept C50644808 @default.
- W4386821167 hasConcept C52622490 @default.
- W4386821167 hasConcept C54170458 @default.
- W4386821167 hasConceptScore W4386821167C131979681 @default.
- W4386821167 hasConceptScore W4386821167C138885662 @default.
- W4386821167 hasConceptScore W4386821167C153180895 @default.
- W4386821167 hasConceptScore W4386821167C154945302 @default.
- W4386821167 hasConceptScore W4386821167C2776151529 @default.
- W4386821167 hasConceptScore W4386821167C2776214188 @default.
- W4386821167 hasConceptScore W4386821167C2776401178 @default.
- W4386821167 hasConceptScore W4386821167C31972630 @default.
- W4386821167 hasConceptScore W4386821167C41008148 @default.
- W4386821167 hasConceptScore W4386821167C41895202 @default.
- W4386821167 hasConceptScore W4386821167C45347329 @default.
- W4386821167 hasConceptScore W4386821167C50644808 @default.
- W4386821167 hasConceptScore W4386821167C52622490 @default.
- W4386821167 hasConceptScore W4386821167C54170458 @default.
- W4386821167 hasLocation W43868211671 @default.
- W4386821167 hasOpenAccess W4386821167 @default.
- W4386821167 hasPrimaryLocation W43868211671 @default.
- W4386821167 hasRelatedWork W1504288058 @default.
- W4386821167 hasRelatedWork W2160739731 @default.
- W4386821167 hasRelatedWork W2546942002 @default.
- W4386821167 hasRelatedWork W2782964878 @default.
- W4386821167 hasRelatedWork W2922421953 @default.
- W4386821167 hasRelatedWork W2979718872 @default.
- W4386821167 hasRelatedWork W3002270006 @default.
- W4386821167 hasRelatedWork W3158534694 @default.
- W4386821167 hasRelatedWork W4317383291 @default.
- W4386821167 hasRelatedWork W4386303287 @default.
- W4386821167 isParatext "false" @default.
- W4386821167 isRetracted "false" @default.
- W4386821167 workType "article" @default.