Electroencephalogram (EEG) plays a pivotal role in the early screening, clinical diagnosis, and prognostic evaluation of neurological disorders. Although EEG-based classification algorithms have achieved remarkable progress in recent years, existing models are primarily designed for static offline scenarios and struggle to adapt to the dynamic characteristics of evolving data distributions over time in clinical settings. While continual learning offers a potential solution, the significant inter-individual variability, non-stationarity, and temporal heterogeneity of EEG signals pose challenges to existing continual learning methods in terms of model adaptability, stability, and the balance between old and new knowledge. To address these issues, this paper proposes a dynamic multi-prototype guided domain-incremental learning method for continual EEG series classification, which employs an evolvable multi-prototype representation guidance mechanism to steer the model. Specifically, we first design a multi-prototype representation strategy that maintains multiple prototypes per class and integrates momentum updates with similarity gating mechanisms to achieve continuous optimization of prototype representations, thereby precisely capturing the dynamic intra-class distribution evolution. Next, we adopt a decoupled training framework for the feature extractor and classifier, leveraging prototype-guided mechanisms to encourage the feature extractor to learn stable inter-task shared representations. Finally, we construct a nearest prototype contrastive loss function to enhance the model’s discriminative capability for decision boundaries and feature structures by optimizing intra-class compactness and inter-class separability. Extensive experimental evaluations on four benchmark datasets demonstrate the effectiveness and efficiency of our proposed method.

This paper presents GRM ($m$), a Gabor wavelet-based face recognition descriptor tailored to tackle the challenges posed by limited-sample conditions in computer vision tasks. Traditional deep learning models, such as ResNet and Transformer architectures, often struggle to generalize with sparse training data, particularly for near-frontal face images. To overcome this limitation, we propose a novel representation framework that leverages Gaussian Riemannian Manifolds (GRM) to capture both geometric structures and statistical dependencies of facial features. The GRM ($m$) descriptor encodes multi-scale local features into a Riemannian manifold space, enhancing the discriminative capability of face representations even with minimal samples. Combined with deep neural networks, GRM ($m$) efficiently fuses handcrafted geometric features with high-level semantic embeddings, significantly improving recognition accuracy. Extensive experiments on benchmark datasets demonstrate that GRM ($m$) outperforms state-of-the-art methods in few-shot learning scenarios, especially under challenging variations in expression, lighting, and accessories. The proposed approach provides a robust and scalable solution for real-world face recognition applications with constrained training samples.

Deep learning models have achieved remarkable progress in precipitation prediction. However, they still face significant challenges in accurately capturing spatial details of radar images, particularly in regions of high precipitation intensity. This limitation results in reduced spatial localization accuracy when predicting radar echo images across varying precipitation intensities. To address this challenge, we propose an innovative precipitation prediction approach termed the Multi- Task Latent Diffusion Model (MTLDM). The core idea of MTLDM lies in the recognition that precipitation radar images represent a combination of multiple components, each corresponding to different precipitation intensities. Thus, we adopt a divide-and-conquer strategy, decomposing radar images into several sub-images based on their precipitation intensities and individually modeling these components. During the prediction stage, MTLDM integrates these sub-image representations by utilizing a trained latent-space rainfall diffusion model, followed by decoding through a multi-task decoder to produce the final precipitation prediction. Experimental evaluations conducted on the MRMS dataset demonstrate that the proposed MTLDM method surpasses state-of-the-art techniques, achieving a Critical Success Index (CSI) improvement of 13-26%.

Deterministic deep learning models for precipitation nowcasting often face several limitations, including cumulative error in long-sequence predictions, over-smoothing, and a reduced ability to capture rare, high-impact rainfall due to data imbalance. To address these challenges, we propose the stagewise scheduled rainfall forecasting network (SSRF-Net), a convolutional framework for continuous multistep rainfall prediction that achieves lower floating-point operations (FLOPs) than competitive baselines under a standardized evaluation. Our framework introduces a multistage, sliding-window prediction mechanism trained with teacher forcing and scheduled sampling to mitigate error accumulation and stabilize training. We design an asymmetric encoder–decoder (E–D) architecture featuring a differential selective encoder (DSE) for selective feature compression and an additive fusion decoder (AFD) that progressively reconstructs details and alleviates over-smoothing. We further introduce an intensity-weighted Gaussian KL divergence loss that aligns sequence-level Gaussian summaries (means and variances) of predictions and ground truth via a KL term, prioritizing heavy-rain events without assuming pixelwise Gaussianity. Extensive experiments on the KNMI and SEVIR datasets show that SSRF-Net outperforms strong baselines, particularly for moderate to severe precipitation; on KNMI, it yields up to 41.8% higher per-frame critical success index (CSI) at the 30-mm/h threshold, with consistent gains on SEVIR.

Deep neural networks have made great achievements in rainfall prediction.However, the current forecasting methods have certain limitations, such as with blurry generated images and incorrect spatial positions. To overcome these challenges, we propose a Two-stage Rainfall-Forecasting Diffusion Model (TRDM) aimed at improving the accuracy of long-term rainfall forecasts and addressing the imbalance in performance between temporal and spatial modeling. TRDM is a two-stage method for rainfall prediction tasks. The task of the first stage is to capture robust temporal information while preserving spatial information under low-resolution conditions. The task of the second stage is to reconstruct the low-resolution images generated in the first stage into high-resolution images. We demonstrate state-of-the-art results on the MRMS and Swedish radar datasets. On the Swedish dataset, our proposed method achieves a 5-10 percentage-point improvement in CSI compared to the other baseline methods for the 60-80 minute prediction range. Our project is open source and available on GitHub at: https://github.com/clearlyzerolxd/TRDM.

Short-term precipitation forecasting remains challenging due to the difficulty in capturing long-term spatiotemporal dependencies. Current deep learning methods fall short in establishing effective dependencies between conditions and forecast results, while also lacking interpretability. To address this issue, we propose a precipitation nowcasting using a diffusion transformer with causal attention (DTCA) model. Our model leverages the transformer and combines causal attention mechanisms to establish spatiotemporal queries between conditional information (causes) and forecast results (results). This design enables the model to effectively capture long-term dependencies, allowing forecast results to maintain strong causal relationships with input conditions over a wide range of time and space. We explore four variants of spatiotemporal information interactions for DTCA, demonstrating that global spatiotemporal labeling interactions yield the best performance. In addition, we introduce a channel-to-batch shift (CTBS) operation to further enhance the model’s ability to represent complex rainfall dynamics. We conducted experiments on two datasets. Compared to state-of-the-art U-Net-based methods, our approach improved the critical success index (CSI) for predicting heavy precipitation by approximately 15% and 8%, respectively, achieving state-of-the-art performance. Our project is open source and available on GitHub at:https://github.com/ybu-lxd/DTCA.

The growing volume of meteorological data and advancements in computing performance have made the application of deep learning technology in short-term rainfall prediction crucial. However, existing learning approaches struggle to accurately predict detailed spatial location information, particularly obvious in predicting extreme rainfall events, leading to inadequate prediction accuracy and subpar performance in meteorological assessment indicators, limiting the effectiveness and applicability of deep learning models in rainfall prediction. To address these challenges, we propose a spacetime separable latent diffusion model with intensity structure information (SSLDM-ISI) to capture spatial and temporal information more efficiently. SSLDM-ISI incorporates two key strategies to solve the spatiotemporal information issue. First, a spatiotemporal conversion block (STC Block) within the backbone network effectively extracts and integrates spatiotemporal information. Second, our proposed latent space coding technique based on rainfall intensity structural information enhances the information representation ability of extreme rainfall. In addition, an examination of the impact of various conditions is conducted on the prediction results to enhance the model’s prediction accuracy and stability. Through comparative analysis of meteorological evaluation and image quality evaluation indicators on two datasets, our proposed approach outperforms existing advanced technologies in short-term rainfall prediction, achieving current state-of-the-art results. Our project is open source and available on GitHub at:https://github.com/ybu-lxd/SISLDM-ISI

We propose Deep Decomposition of Circularly Symmetric Gabor Wavelet (DD-CSGW) for rotation-invariant texture image classification. Circularly Symmetric Gabor Wavelet (CSGW) is rotation-invariant tool for image analysis. However, CSGW has an obvious shortcoming: it extracts less discriminative information from image due to lack of directional selectivity. We make two contributions to improve the performance of CSGW: (1) We propose deep decomposition approach of CSGW to obtain more discriminative image information; (2) Because strong scale dependencies exist in the domain of CSGW, we use copula model to capture these scale dependencies. For classification, the energies of DD-CSGW and the parameters of copula model based on DD-CSGW are used as the features of texture, and SVM is utilized as the classifier. Experiments show that DD-CSGW obviously improves the performance of CSGW, and it is effective compared with the state-of-the-art rotation-invariant methods.

Obtaining robust and efficient rotation-invariant texture features in content-based image retrieval field is a challenging work. We propose three efficient rotation-invariant methods for texture image retrieval using copula model based in the domains of Gabor wavelet (GW) and circularly symmetric GW (CSGW). The proposed copula models use copula function to capture the scale dependence of GW/CSGW for improving the retrieval performance. It is well known that the Kullback-Leibler distance (KLD) is the commonly used similarity measurement between probability models. However, it is difficult to deduce the closed-form of KLD between two copula models due to the complexity of the copula model. We also put forward a kind of retrieval scheme using the KLDs of marginal distributions and the KLD of copula function to calculate the KLD of copula model. The proposed texture retrieval method has low computational complexity and high retrieval precision. The experimental results on VisTex and Brodatz data sets show that the proposed retrieval method is more effective compared with the state-of-the-art methods.