The block-based coding structure in the hybrid video framework inevitably introduces compression artifacts such as blocking, ringing, and so on. To compensate for those artifacts, extensive filtering techniques were proposed loop of codecs, which are capable boosting subjective objective qualities reconstructed videos. Recently, neural network-based filters presented with power deep learning fr...

Traffic accidents directly influence public safety and economic development; thus, the prevention of traffic is great importance in urban transportation. The accurate prediction can assist departments to better control prevent accidents. Thus, this paper proposes a deep learning method named attention-based residual dilated network (ARDN), extract essential information from multi-source dataset...

The pooling operation is used in graph classification tasks to leverage hierarchical structures preserved data and reduce computational complexity. However, shrinkage discards details, existing methods may lead the loss of key features. In this work, we propose a residual convolutional neural network tackle problem features losing. Particularly, our contributions are threefold: (1) Different fr...

Multi-behavior recommendation exploits multiple types of user-item interactions, such as view and cart , to learn user preferences has demonstrated be an effective solution alleviate the data sparsity problem faced by traditional models that often utilize only one type interaction for recommendation. In real scenarios, users take a sequence actions interact with item, in order get more informat...

Among the most crucial methods for denoising a noisy voice signal and enhancing its quality is speech enhancement. This study makes use of Adaptive Residual Neural Network technique to reduces maximum off background noise. method continuously monitors noise depends upon environmental changes using SNR parameter. It has two functions first one non linear followed by convolutional neural networks...