The effectiveness of modeling contextual information has been empirically shown in numerous computer vision tasks. In this paper, we propose a simple yet efficient augmented fully convolutional network(AugFCN) by aggr...
详细信息
The effectiveness of modeling contextual information has been empirically shown in numerous computer vision tasks. In this paper, we propose a simple yet efficient augmented fully convolutional network(AugFCN) by aggregating content-and position-based object contexts for semantic ***, motivated because each deep feature map is a global, class-wise representation of the input,we first propose an augmented nonlocal interaction(AugNI) to aggregate the global content-based contexts through all feature map interactions. Compared to classical position-wise approaches, AugNI is more efficient. Moreover, to eliminate permutation equivariance and maintain translation equivariance, a learnable,relative position embedding branch is then supportably installed in AugNI to capture the global positionbased contexts. AugFCN is built on a fully convolutional network as the backbone by deploying AugNI before the segmentation head network. Experimental results on two challenging benchmarks verify that AugFCN can achieve a competitive 45.38% mIoU(standard mean intersection over union) and 81.9% mIoU on the ADE20K val set and Cityscapes test set, respectively, with little computational overhead. Additionally, the results of the joint implementation of AugNI and existing context modeling schemes show that AugFCN leads to continuous segmentation improvements in state-of-the-art context modeling. We finally achieve a top performance of 45.43% mIoU on the ADE20K val set and 83.0% mIoU on the Cityscapes test set.
Foundation models(FMs) [1] have revolutionized software development and become the core components of large software systems. This paradigm shift, however, demands fundamental re-imagining of software engineering theo...
Foundation models(FMs) [1] have revolutionized software development and become the core components of large software systems. This paradigm shift, however, demands fundamental re-imagining of software engineering theories and methodologies [2]. Instead of replacing existing software modules implemented by symbolic logic, incorporating FMs' capabilities to build software systems requires entirely new modules that leverage the unique capabilities of ***, while FMs excel at handling uncertainty, recognizing patterns, and processing unstructured data, we need new engineering theories that support the paradigm shift from explicitly programming and maintaining user-defined symbolic logic to creating rich, expressive requirements that FMs can accurately perceive and implement.
Purpose-The Internet of Things(IoT)cloud platforms provide end-to-end solutions that integrate various capabilities such as application development,device and connectivity management,data storage,data analysis and dat...
详细信息
Purpose-The Internet of Things(IoT)cloud platforms provide end-to-end solutions that integrate various capabilities such as application development,device and connectivity management,data storage,data analysis and data *** high use of these platforms results in their huge availability provided by different ***,choosing the optimal IoT cloud platform to develop IoT applications successfully has become *** key purpose of the present study is to implement a hybrid multi-attribute decision-making approach(MADM)to evaluate and select IoT cloud ***/methodology/approach-The optimal selection of the IoT cloud platforms seems to be dependent on multiple ***,the optimal selection of IoT cloud platforms problem is modeled as a MADM problem,and a hybrid approach named neutrosophic fuzzy set-Euclidean taxicab distance-based approach(NFS-ETDBA)is implemented to solve the ***-ETDBA works on the calculation of assessment score for each alternative,*** cloud platforms,by combining two different measures:Euclidean and taxicab ***-A case study to illustrate the working of the proposed NFS-ETDBA for optimal selection of IoT cloud platforms is *** results obtained on the basis of calculated assessment scores depict that“Azure IoT suite”is the most preferable IoT cloud platform,whereas“Salesman IoT cloud”is the least ***/value-The proposed NFS-ETDBA methodology for the IoT cloud platform selection is implemented for the first time in this *** is highly capable of handling the large number of alternatives and the selection attributes involved in any decision-making ***,the use of fuzzy set theory(FST)makes it very easy to handle the impreciseness that may occur during the data collection through a questionnaire from a group of experts.
Over the past two decades, the rise in video streaming has been driven by internet accessibility and the demand for high-quality video. To meet this demand across varying network speeds and devices, transcoding is ess...
详细信息
The Hungarian algorithm is a well-known cubic-time algorithm for finding minimum-cost matchings in weighted bipartite graphs. While utilizing it for multi-agent path planning yields the minimum-total-length set of pat...
详细信息
Trojan detection from network traffic data is crucial for safeguarding networks against covert infiltration and potential data breaches. Deep learning (DL) techniques can play a pivotal role in detecting trojans from ...
详细信息
In the realm of deep learning, the prevalence of models with large number of parameters poses a significant challenge for low computation device. Critical influence of model size, primarily governed by weight paramete...
详细信息
In this paper, we delve into the transformative landscape of education amidst the disruptive advances of generative AI (GenAI), characterized by an unprecedented capacity to generate new information with tools such as...
详细信息
Long-term urban traffic flow prediction is an important task in the field of intelligent transportation,as it can help optimize traffic management and improve travel *** improve prediction accuracy,a crucial issue is ...
详细信息
Long-term urban traffic flow prediction is an important task in the field of intelligent transportation,as it can help optimize traffic management and improve travel *** improve prediction accuracy,a crucial issue is how to model spatiotemporal dependency in urban traffic *** recent years,many studies have adopted spatiotemporal neural networks to extract key information from traffic ***,most models ignore the semantic spatial similarity between long-distance areas when mining spatial *** also ignore the impact of predicted time steps on the next unpredicted time step for making long-term ***,these models lack a comprehensive data embedding process to represent complex spatiotemporal *** paper proposes a multi-scale persistent spatiotemporal transformer(MSPSTT)model to perform accurate long-term traffic flow prediction in *** adopts an encoder-decoder structure and incorporates temporal,periodic,and spatial features to fully embed urban traffic data to address these *** model consists of a spatiotemporal encoder and a spatiotemporal decoder,which rely on temporal,geospatial,and semantic space multi-head attention modules to dynamically extract temporal,geospatial,and semantic *** spatiotemporal decoder combines the context information provided by the encoder,integrates the predicted time step information,and is iteratively updated to learn the correlation between different time steps in the broader time range to improve the model’s accuracy for long-term *** on four public transportation datasets demonstrate that MSPSTT outperforms the existing models by up to 9.5%on three common metrics.
Near-field tides prediction for tsunami detection in the coastal area is a significant problem of the cable-based tsunami meter system in north Sipora, Indonesia. The problem is caused by its shallow water condition a...
详细信息
暂无评论