Fully-convolutional Siamese networks for visual tracking have drawn great attention in balancing tracking accuracy and speed. However, there is still some inherent inaccuracy with advanced trackers, since they only learn a general matching model from large scale datasets by off-line training. This generates the target template without sufficient discriminant information and does not adapt well to the current tracking sequence. In this paper, we introduce the channel attention mechanism into the network to better learn the matching model and, during the online tracking phase, we design an initial matting guidance strategy in which: 1) the superpixel matting algorithm is applied to extract the target foreground in the initial frame, and 2) the matted image with foreground only is fed into the network and fused with the original image feature. Under matting guidance, the fused target template has more details for representation of target appearance and more structural information from superpixels for robust tracking. The experimental results on object tracking benchmark (OTB) show that our approach achieves excellent performance while it also provides real-time tracking speed. INDEX TERMSVisual tracking, siamese network, matching model, superpixel, channel attention.

No abstract

No abstract

Medical image segmentation is a fundamental task in medical image analysis. Dynamic receptive field is very helpful for accurate medical image segmentation, which needs to be further studied and utilized. In this paper, we propose Match Feature U-Net, a novel, symmetric encoder– decoder architecture with dynamic receptive field for medical image segmentation. We modify the Selective Kernel convolution (a module proposed in Selective Kernel Networks) by inserting a newly proposed Match operation, which makes similar features in different convolution branches have corresponding positions, and then we replace the U-Net’s convolution with the redesigned Selective Kernel convolution. This network is a combination of U-Net and improved Selective Kernel convolution. It inherits the advantages of simple structure and low parameter complexity of U-Net, and enhances the efficiency of dynamic receptive field in Selective Kernel convolution, making it an ideal model for medical image segmentation tasks which often have small training data and large changes in targets size. Compared with state-of-the-art segmentation methods, the number of parameters in Match Feature U-Net (2.65 M) is 34% of U-Net (7.76 M), 29% of UNet++ (9.04 M), and 9.1% of CE-Net (29 M). We evaluated the proposed architecture in four medical image segmentation tasks: nuclei segmentation in microscopy images, breast cancer cell segmentation, gland segmentation in colon histology images, and disc/cup segmentation. Our experimental results show that Match Feature U-Net achieves an average Mean Intersection over Union (MIoU) gain of 1.8, 1.45, and 2.82 points over U-Net, UNet++, and CE-Net, respectively.

Background Protein–protein interactions are widespread in biological systems and play an important role in cell biology. Since traditional laboratory-based methods have some drawbacks, such as time-consuming, money-consuming, etc., a large number of methods based on deep learning have emerged. However, these methods do not take into account the long-distance dependency information between each two amino acids in sequence. In addition, most existing models based on graph neural networks only aggregate the first-order neighbors in protein–protein interaction (PPI) network. Although multi-order neighbor information can be aggregated by increasing the number of layers of neural network, it is easy to cause over-fitting. So, it is necessary to design a network that can capture long distance dependency information between amino acids in the sequence and can directly capture multi-order neighbor information in protein–protein interaction network. Results In this study, we propose a multi-hop neural network (LDMGNN) model combining long distance dependency information to predict the multi-label protein–protein interactions. In the LDMGNN model, we design the protein amino acid sequence encoding (PAASE) module with the multi-head self-attention Transformer block to extract the features of amino acid sequences by calculating the interdependence between every two amino acids. And expand the receptive field in space by constructing a two-hop protein–protein interaction (THPPI) network. We combine PPI network and THPPI network with amino acid sequence features respectively, then input them into two identical GIN blocks at the same time to obtain two embeddings. Next, the two embeddings are fused and input to the classifier for predict multi-label protein–protein interactions. Compared with other state-of-the-art methods, LDMGNN shows the best performance on both the SHS27K and SHS148k datasets. Ablation experiments show that the PAASE module and the construction of THPPI network are feasible and effective. Conclusions In general terms, our proposed LDMGNN model has achieved satisfactory results in the prediction of multi-label protein–protein interactions.

There has been significant progress in skeleton-based action recognition. Human skeleton can be naturally structured into graph, so graph convolution networks have become the most popular method in this task. Most of these state-of-the-art methods optimized the structure of human skeleton graph to obtain better performance. Based on these advanced algorithms, a simple but strong network is proposed with three major contributions. Firstly, inspired by some adaptive graph convolution networks and non-local blocks, some kinds of self-attention modules are designed to exploit spatial and temporal dependencies and dynamically optimize the graph structure. Secondly, a light but efficient architecture of network is designed for skeleton-based action recognition. Moreover, a trick is proposed to enrich the skeleton data with bones connection information and make obvious improvement to the performance. The method achieves 90.5% accuracy on cross-subjects setting (NTU60), with 0.89M parameters and 0.32 GMACs of computation cost. This work is expected to inspire new ideas for the field.

No abstract

No abstract