As an aperture-divided computational imaging system, microlens array (MLA) -based light field (LF) imaging is playing an increasingly important role in computer vision. As the trade-off between the spatial and angular resolutions, deep learning (DL) -based image super-resolution (SR) methods have been applied to enhance the spatial resolution. However, in existing DL-based methods, the depth-varying defocus is not considered both in dataset development and algorithm design, which restricts many applications such as depth estimation and object recognition. To overcome this shortcoming, a super-resolution task that reconstructs all-in-focus high-resolution (HR) LF images from low-resolution (LR) LF images is proposed by designing a large dataset and proposing a convolutional neural network (CNN) -based SR method. The dataset is constructed by using Blender software, consisting of 150 light field images used as training data, and 15 light field images used as validation and testing data. The proposed network is designed by proposing the dilated deformable convolutional network (DCN) -based feature extraction block and the LF subaperture image (SAI) Deblur-SR block. The experimental results demonstrate that the proposed method achieves more appealing results both quantitatively and qualitatively.

As human actions can be characterized by the trajectories of skeleton joints, skeleton-based action recognition techniques have gained increasing attention in the field of intelligent recognition and behavior analysis. With the emergence of large datasets, graph convolutional network (GCN) approaches have been widely applied for skeleton-based action recognition and have achieved remarkable performances. In this paper, a novel GCN-based approach is proposed by introducing a convolutional block attention module (CBAM)-based graph attention block to compute the semantic correlations between any two vertices. By considering semantic correlations, our model can effectively identify the most discriminative vertex connections associated with specific actions, even when the two vertices are physically unconnected. Experimental results demonstrate that the proposed model is effective and outperforms existing methods.

Modern city construction focuses on developing smart transportation, but the recognition of the large number of non-motorized vehicles in the city is still not sufficient. Compared to fixed recognition equipment, drones have advantages in image acquisition due to their flexibility and maneuverability. With the dataset collected from aerial images taken by drones, this study proposed a novel lightweight architecture for small objection detection based on YOLO framework, named EBR-YOLO. Firstly, since the targets in the application scenario are generally small, the number of Backbone layers is reduced, and the AZML module is proposed to enrich the detail information and enhance the model learning capability. Secondly, the C2f module is reconstructed using part of the convolutional PConv to reduce the network’s computational volume and improve the detection speed. Finally, the downsampling operation is reshaped by combining with the introduced ADown module to further reduce the computational amount of the model. The experimental results show that the algorithm achieves an mAP of 98.9% and an FPS of 89.8 on the self-built dataset of this paper, which is only 0.2% and 0.3 lower compared to the original YOLOv8 network, respectively, and the number of parameters is 70% lower compared to the baseline, which ensures the accuracy and computational speed of the model while reducing its computational volume greatly. At the same time, the model generalization experiments are carried out on the UCAS-AOD and CARPK datasets, and the performance of the model is almost the same as the baseline.

Student–teacher networks have shown promise in unsupervised anomaly detection; however, issues such as semantic confusion and abnormal deformations still restrict the detection accuracy. To address these issues, we propose a novel student–teacher network named MST by integrating the multistage pixel-reserving bridge (MPRB) and the spatial compression autoencoder (SCA) to the MMR network. The MPRB enhances inter-level information interaction and local feature extraction, improving the anomaly localization and reducing the false detection area. The SCA bolsters global feature extraction, making the detection boundaries of larger defects clearer. By testing our network across various datasets, our method achieves state-of-the-art (SOTA) performance on AeBAD-S, AeBAD-V, and MPDD datasets, with image-level AUROC scores of 87.5%, 78.5%, and 96.5%, respectively. Furthermore, our method also exhibits competitive performance on the widely utilized MVTec AD dataset.