Few-shot semantic segmentation aims at training a model that can segment novel classes in a query image with only a few densely annotated support exemplars.It remains a challenge because of large intra-class variation...Few-shot semantic segmentation aims at training a model that can segment novel classes in a query image with only a few densely annotated support exemplars.It remains a challenge because of large intra-class variations between the support and query images.Existing approaches utilize 4D convolutions to mine semantic correspondence between the support and query images.However,they still suffer from heavy computation,sparse correspondence,and large memory.We propose axial assembled correspondence network(AACNet)to alleviate these issues.The key point of AACNet is the proposed axial assembled 4D kernel,which constructs the basic block for semantic correspondence encoder(SCE).Furthermore,we propose the deblurring equations to provide more robust correspondence for the aforementioned SCE and design a novel fusion module to mix correspondences in a learnable manner.Experiments on PASCAL-5~i reveal that our AACNet achieves a mean intersection-over-union score of 65.9%for 1-shot segmentation and 70.6%for 5-shot segmentation,surpassing the state-of-the-art method by 5.8%and 5.0%respectively.展开更多
A composite rubber concrete(CRC)was designed by combining waste tire rubber particles with particle sizes of 3~5 mm,1~3 mm and 20 mesh.Taking the rubber content of different particle sizes as the influencing factors,t...A composite rubber concrete(CRC)was designed by combining waste tire rubber particles with particle sizes of 3~5 mm,1~3 mm and 20 mesh.Taking the rubber content of different particle sizes as the influencing factors,the range and variance analysis of the mechanical and impermeability properties of CRC was carried out by orthogonal test.Through analysis,it is concluded that the optimal proportion of 3~5 mm,1~3 mm,and 20 mesh particle size composite rubber is 1:2.5:5.5 kinds of CRC and 3 kinds of ordinary single-mixed rubber concrete(RC)with a total content of 10%~20%were designed under this ratio,and the salt-freezing cycle test was carried out with a concentration of 5%Na 2 SO4 solution.The physical and mechanical damage laws during 120 salt-freezing cycles are obtained,and the corresponding damage prediction model is established according to the experimental data.The results show that:on the one hand,the composite rubber in CRC produces a more uniform“graded”structure,forms a retractable particle group,and reduces the loss of mechanical properties of CRC.On the other hand,colloidal particles with different particle sizes are used as air entraining agent to improve the pore structure of concrete and introduce evenly dispersed bubbles,which fundamentally improves the durability of concrete.Under the experimental conditions,the CRC performance is the best when the overall content of composite rubber is 15%.展开更多
基金supported in part by the Key Research and Development Program of Guangdong Province(2021B0101200001)the Guangdong Basic and Applied Basic Research Foundation(2020B1515120071)。
文摘Few-shot semantic segmentation aims at training a model that can segment novel classes in a query image with only a few densely annotated support exemplars.It remains a challenge because of large intra-class variations between the support and query images.Existing approaches utilize 4D convolutions to mine semantic correspondence between the support and query images.However,they still suffer from heavy computation,sparse correspondence,and large memory.We propose axial assembled correspondence network(AACNet)to alleviate these issues.The key point of AACNet is the proposed axial assembled 4D kernel,which constructs the basic block for semantic correspondence encoder(SCE).Furthermore,we propose the deblurring equations to provide more robust correspondence for the aforementioned SCE and design a novel fusion module to mix correspondences in a learnable manner.Experiments on PASCAL-5~i reveal that our AACNet achieves a mean intersection-over-union score of 65.9%for 1-shot segmentation and 70.6%for 5-shot segmentation,surpassing the state-of-the-art method by 5.8%and 5.0%respectively.
基金supported by the National Key Research and Development Program of China under the Grant No.2018YFC0809400.
文摘A composite rubber concrete(CRC)was designed by combining waste tire rubber particles with particle sizes of 3~5 mm,1~3 mm and 20 mesh.Taking the rubber content of different particle sizes as the influencing factors,the range and variance analysis of the mechanical and impermeability properties of CRC was carried out by orthogonal test.Through analysis,it is concluded that the optimal proportion of 3~5 mm,1~3 mm,and 20 mesh particle size composite rubber is 1:2.5:5.5 kinds of CRC and 3 kinds of ordinary single-mixed rubber concrete(RC)with a total content of 10%~20%were designed under this ratio,and the salt-freezing cycle test was carried out with a concentration of 5%Na 2 SO4 solution.The physical and mechanical damage laws during 120 salt-freezing cycles are obtained,and the corresponding damage prediction model is established according to the experimental data.The results show that:on the one hand,the composite rubber in CRC produces a more uniform“graded”structure,forms a retractable particle group,and reduces the loss of mechanical properties of CRC.On the other hand,colloidal particles with different particle sizes are used as air entraining agent to improve the pore structure of concrete and introduce evenly dispersed bubbles,which fundamentally improves the durability of concrete.Under the experimental conditions,the CRC performance is the best when the overall content of composite rubber is 15%.