With the popularity of using deep learning-based models in various categorization problems and their proven robustness compared to conventional methods, a growing number of researchers have exploited such methods in environment sound classification tasks in recent years. However, the performances of existing models use auditory features like log-mel spectrogram (LM) and mel frequency cepstral coefficient (MFCC), or raw waveform to train deep neural networks for environment sound classification (ESC) are unsatisfactory. In this paper, we first propose two combined features to give a more comprehensive representation of environment sounds Then, a fourfour-layer convolutional neural network (CNN) is presented to improve the performance of ESC with the proposed aggregated features. Finally, the CNN trained with different features are fused using the Dempster–Shafer evidence theory to compose TSCNN-DS model. The experiment results indicate that our combined features with the four-layer CNN are appropriate for environment sound taxonomic problems and dramatically outperform other conventional methods. The proposed TSCNN-DS model achieves a classification accuracy of 97.2%, which is the highest taxonomic accuracy on UrbanSound8K datasets compared to existing models.

In the past few years, visual tracking methods with convolution neural networks (CNNs) have gained great popularity and success. However, the convolution operation of CNNs struggles to relate spatially distant information, which limits the discriminative power of trackers. Very recently, several Transformer-assisted tracking approaches have emerged to alleviate the above issue by combining CNNs with Transformers to enhance the feature representation. In contrast to the methods mentioned above, this article explores a pure Transformer-based model with a novel semi-Siamese architecture. Both the time–space self-attention module used to construct the feature extraction backbone and the cross-attention discriminator used to estimate the response map solely leverage attention without convolution. Inspired by the recent vision transformers (ViTs), we propose the multistage alternating time–space Transformers (ATSTs) to learn robust feature representation. Specifically, temporal and spatial tokens at each stage are alternately extracted and encoded by separate Transformers. Subsequently, a cross-attention discriminator is proposed to directly generate response maps of the search region without additional prediction heads or correlation filters. Experimental results show that our ATST-based model attains favorable results against state-of-the-art convolutional trackers. Moreover, it shows comparable performance with recent “CNN + Transformer” trackers on various benchmarks while our ATST requires significantly less training data.

Deep learning-based segmentation methods have demonstrated significant performance over their traditional counterparts. However, striving for better accuracy with such networks usually leads to the deterioration of the network’s computational efficiency, thereby rendering them inefficient for deployment on resource constraint devices. Establishing the required tradeoff between the accuracy of pixel prediction and computational efficiency remains challenging. In this article, a lightweight multiscale segmentation framework is proposed. We leverage the representation power of different receptive fields to attain optimal accuracy while maintaining computational efficiency by embedding the sparse network architecture with the depthwise separable convolution at the multiscale level. Experimental results from two challenging remote sensing segmentation datasets show that the proposed network can achieve substantial pixel prediction accuracy at relatively low computational overhead compared to state-of-the-art networks.

Video inpainting aims to complete missing frames visually convincingly by balancing high-frequency detailed textures and low-frequency semantic structures. Conventional approaches utilize generative adversarial and reconstruction losses for optimizing output frames, each favoring different frequency aspects, to achieve this equilibrium. However, employing both loss types concurrently often results in a conflict between perceptual and distortion qualities, mainly due to their distinct frequency preferences. In response, this letter introduces the Waveletbased Transformer network for Video Inpainting (WTVI). WTVI employs a 2D discrete wavelet transform (DWT) to decompose frames into various frequency bands, ensuring the preservation of spatial information. It then independently completes missing regions in each band using Transformer network. To mitigate inter-frequency conflicts, we apply reconstruction loss to the low-frequency bands and adversarial loss to the high-frequency bzands. Additionally, we innovate High-frequency Cross-Attention (HCA) and Low-frequency Cross-Attention (LCA) modules to enhance frequency dependency learning beyond the spatialtemporal scope and to align features across bands. Our experiments confirm that WTVI surpasses previous methods, significantly improving both quantitative and qualitative performance.