In recent years,semantic segmentation on 3D point cloud data has attracted much attention.Unlike 2D images where pixels distribute regularly in the image domain,3D point clouds in non-Euclidean space are irregular and...In recent years,semantic segmentation on 3D point cloud data has attracted much attention.Unlike 2D images where pixels distribute regularly in the image domain,3D point clouds in non-Euclidean space are irregular and inherently sparse.Therefore,it is very difficult to extract long-range contexts and effectively aggregate local features for semantic segmentation in 3D point cloud space.Most current methods either focus on local feature aggregation or long-range context dependency,but fail to directly establish a global-local feature extractor to complete the point cloud semantic segmentation tasks.In this paper,we propose a Transformer-based stratified graph convolutional network(SGT-Net),which enlarges the effective receptive field and builds direct long-range dependency.Specifically,we first propose a novel dense-sparse sampling strategy that provides dense local vertices and sparse long-distance vertices for subsequent graph convolutional network(GCN).Secondly,we propose a multi-key self-attention mechanism based on the Transformer to further weight augmentation for crucial neighboring relationships and enlarge the effective receptive field.In addition,to further improve the efficiency of the network,we propose a similarity measurement module to determine whether the neighborhood near the center point is effective.We demonstrate the validity and superiority of our method on the S3DIS and ShapeNet datasets.Through ablation experiments and segmentation visualization,we verify that the SGT model can improve the performance of the point cloud semantic segmentation.展开更多
Deep learning, especially through convolutional neural networks (CNN) such as the U-Net 3D model, has revolutionized fault identification from seismic data, representing a significant leap over traditional methods. Ou...Deep learning, especially through convolutional neural networks (CNN) such as the U-Net 3D model, has revolutionized fault identification from seismic data, representing a significant leap over traditional methods. Our review traces the evolution of CNN, emphasizing the adaptation and capabilities of the U-Net 3D model in automating seismic fault delineation with unprecedented accuracy. We find: 1) The transition from basic neural networks to sophisticated CNN has enabled remarkable advancements in image recognition, which are directly applicable to analyzing seismic data. The U-Net 3D model, with its innovative architecture, exemplifies this progress by providing a method for detailed and accurate fault detection with reduced manual interpretation bias. 2) The U-Net 3D model has demonstrated its superiority over traditional fault identification methods in several key areas: it has enhanced interpretation accuracy, increased operational efficiency, and reduced the subjectivity of manual methods. 3) Despite these achievements, challenges such as the need for effective data preprocessing, acquisition of high-quality annotated datasets, and achieving model generalization across different geological conditions remain. Future research should therefore focus on developing more complex network architectures and innovative training strategies to refine fault identification performance further. Our findings confirm the transformative potential of deep learning, particularly CNN like the U-Net 3D model, in geosciences, advocating for its broader integration to revolutionize geological exploration and seismic analysis.展开更多
Photocatalytic H_(2) evolution from seawater splitting presents a promising approach to tackle the fossil energy crisis and mitigate carbon emission due to the abundant source of seawater and sunlight on the earth.How...Photocatalytic H_(2) evolution from seawater splitting presents a promising approach to tackle the fossil energy crisis and mitigate carbon emission due to the abundant source of seawater and sunlight on the earth.However,the development of efficient photocatalysts for seawater splitting remains a formidable challenge.Herein,a 2D/2D ZnIn_(2)S_(4)/WO_(3)(ZIS/WO_(3))heterojunction nanostructure is fabricated to efficiently separate the photoinduced carriers by steering electron transfer from the conduction band minimum of WO_(3) to the valence band maximum of ZIS via constructing internal electric field.Subsequently,plasmonic Au nanoparticles(NPs)as a novel photosensitizer and a reduction cocatalyst are anchored on ZIS/WO_(3) surface to further enhance the optical absorption of ZIS/WO_(3) heterojunction and accelerate the catalytic conversion.The obtained Au/ZIS/WO_(3) photocatalyst exhibits an outstanding H_(2) evolution rate of 2610.6 or 3566.3μmol g^(-1)h~(-1)from seawater splitting under visible or full-spectrum light irradiation,respectively.These rates represent an impressive increase of approximately 7.3-and 6,6-fold compared to those of ZIS under the illumination of the same light source.The unique 2D/2D structure,internal electric field,and plasmonic metal modification together boost the photocatalytic H_(2) evolution rate of Au/ZIS/WO_(3),making it even comparable to H_(2) evolution from pure water splitting.The present work sheds light on the development of efficient photocatalysts for seawater splitting.展开更多
基金supported in part by the National Natural Science Foundation of China under Grant Nos.U20A20197,62306187the Foundation of Ministry of Industry and Information Technology TC220H05X-04.
文摘In recent years,semantic segmentation on 3D point cloud data has attracted much attention.Unlike 2D images where pixels distribute regularly in the image domain,3D point clouds in non-Euclidean space are irregular and inherently sparse.Therefore,it is very difficult to extract long-range contexts and effectively aggregate local features for semantic segmentation in 3D point cloud space.Most current methods either focus on local feature aggregation or long-range context dependency,but fail to directly establish a global-local feature extractor to complete the point cloud semantic segmentation tasks.In this paper,we propose a Transformer-based stratified graph convolutional network(SGT-Net),which enlarges the effective receptive field and builds direct long-range dependency.Specifically,we first propose a novel dense-sparse sampling strategy that provides dense local vertices and sparse long-distance vertices for subsequent graph convolutional network(GCN).Secondly,we propose a multi-key self-attention mechanism based on the Transformer to further weight augmentation for crucial neighboring relationships and enlarge the effective receptive field.In addition,to further improve the efficiency of the network,we propose a similarity measurement module to determine whether the neighborhood near the center point is effective.We demonstrate the validity and superiority of our method on the S3DIS and ShapeNet datasets.Through ablation experiments and segmentation visualization,we verify that the SGT model can improve the performance of the point cloud semantic segmentation.
文摘Deep learning, especially through convolutional neural networks (CNN) such as the U-Net 3D model, has revolutionized fault identification from seismic data, representing a significant leap over traditional methods. Our review traces the evolution of CNN, emphasizing the adaptation and capabilities of the U-Net 3D model in automating seismic fault delineation with unprecedented accuracy. We find: 1) The transition from basic neural networks to sophisticated CNN has enabled remarkable advancements in image recognition, which are directly applicable to analyzing seismic data. The U-Net 3D model, with its innovative architecture, exemplifies this progress by providing a method for detailed and accurate fault detection with reduced manual interpretation bias. 2) The U-Net 3D model has demonstrated its superiority over traditional fault identification methods in several key areas: it has enhanced interpretation accuracy, increased operational efficiency, and reduced the subjectivity of manual methods. 3) Despite these achievements, challenges such as the need for effective data preprocessing, acquisition of high-quality annotated datasets, and achieving model generalization across different geological conditions remain. Future research should therefore focus on developing more complex network architectures and innovative training strategies to refine fault identification performance further. Our findings confirm the transformative potential of deep learning, particularly CNN like the U-Net 3D model, in geosciences, advocating for its broader integration to revolutionize geological exploration and seismic analysis.
基金supported by the National Natural Science Foundation of China(21872104,21501131,21978216 and 22272082)the Natural Science Foundation of Tianjin for Distinguished Young Scholar(20JCJQJC00150)the Analytical&Testing Center of Tiangong University for PL work。
文摘Photocatalytic H_(2) evolution from seawater splitting presents a promising approach to tackle the fossil energy crisis and mitigate carbon emission due to the abundant source of seawater and sunlight on the earth.However,the development of efficient photocatalysts for seawater splitting remains a formidable challenge.Herein,a 2D/2D ZnIn_(2)S_(4)/WO_(3)(ZIS/WO_(3))heterojunction nanostructure is fabricated to efficiently separate the photoinduced carriers by steering electron transfer from the conduction band minimum of WO_(3) to the valence band maximum of ZIS via constructing internal electric field.Subsequently,plasmonic Au nanoparticles(NPs)as a novel photosensitizer and a reduction cocatalyst are anchored on ZIS/WO_(3) surface to further enhance the optical absorption of ZIS/WO_(3) heterojunction and accelerate the catalytic conversion.The obtained Au/ZIS/WO_(3) photocatalyst exhibits an outstanding H_(2) evolution rate of 2610.6 or 3566.3μmol g^(-1)h~(-1)from seawater splitting under visible or full-spectrum light irradiation,respectively.These rates represent an impressive increase of approximately 7.3-and 6,6-fold compared to those of ZIS under the illumination of the same light source.The unique 2D/2D structure,internal electric field,and plasmonic metal modification together boost the photocatalytic H_(2) evolution rate of Au/ZIS/WO_(3),making it even comparable to H_(2) evolution from pure water splitting.The present work sheds light on the development of efficient photocatalysts for seawater splitting.