Music emotion recognition is a key task in symbolic music understanding (SMER). Recent approaches have shown promising results by fine-tuning large-scale pre-trained models (e.g., MIDIBERT, a benchmark in symbolic music understanding) to map musical semantics to emotional labels. While these models effectively capture distributional musical semantics, they often overlook tonal structures, particularly musical modes, which play a critical role in emotional perception according to music psychology. In this paper, we investigate the representational capacity of MIDIBERT and identify its limitations in capturing mode-emotion associations. To address this issue, we propose a Mode-Guided Enhancement (MoGE) strategy that incorporates psychological insights on mode into the model. Specifically, we first conduct a mode augmentation analysis, which reveals that MIDIBERT fails to effectively encode emotion-mode correlations. Motivated by this observation, we further identify the MIDIBERT layer that shows the weakest emotion relevance and introduce a Mode-guided Feature-wise linear modulation injection (MoFi) framework to inject explicit mode features, thereby enhancing the model's capability in emotional representation and inference. Extensive experiments on the EMOPIA and VGMIDI datasets demonstrate that our mode injection strategy significantly improves SMER performance, achieving accuracies of 75.2% and 59.1%, respectively. These results validate the effectiveness of mode-guided modeling in symbolic music emotion recognition.

Understanding motion is essential for visual object tracking, especially in complex and dynamic scenarios. Yet, many existing methods rely on simplistic strategies such as template updates or temporal feature propagation, often overlooking the deeper modeling of motion information. To mitigate this limitation, we introduce a motion-aware spatio-temporal framework that enhances motion perception by explicitly matching motion patterns and modeling inter-frame motion relationships. Central to our design is a motion pattern dictionary, which encodes a diverse set of representative motion cues as learnable features. During tracking, features from the search region interact with the dictionary to retrieve the most relevant motion patterns, allowing the model to adapt to the current motion state. A dedicated decoder further incorporates temporal correlations to refine motion awareness. To complement motion modeling, we embed geometric cues into the search region features, which strengthens spatial perception, reduces ambiguity under occlusion, and improves foreground-background separation. Extensive evaluations on seven challenging benchmarks demonstrate the effectiveness of our design. In particular, MoDTrack_384 surpasses recent SOTA trackers on LaSOT by 1.2% in AUC, highlighting the benefits of motion pattern modeling and geometry-guided enhancement in mitigating tracking drift.

Existing Optical Remote Sensing Image Salient Object Detection (ORSI-SOD) methods mainly rely on a semantic segmentation paradigm, which relies on pixel-wise probabilities, leading to overconfident mispredictions. In contrast, the random sampling process of the diffusion model allows multiple possible predictions to be drawn from the mask distribution, effectively alleviating this problem. However, existing diffusion models mainly use Transformers as conditional feature extraction networks. Although they are good at global modeling, they have limited ability to handle long-range dependencies due to computational complexity. To overcome these challenges, we introduce MambaDif, an innovative diffusion model architecture based on Mamba. Specifically, we regard ORSI-SOD as a conditional mask generation task leveraging the diffusion model and achieving target distribution matching by adding noise to the mask and iteratively denoising it to match the target distribution. Then, we adopt Mamba to extract global features, efficiently process long sequences, and capture global contextual information with linear complexity. In addition, we introduce the global-local feature collaborative completion module (GLM), which combines the ability of convolutional layers to extract local features with the advantage of Mamba in capturing long-range dependencies, thereby achieving excellent denoising performance. Extensive experiments show that MambaDif outperforms SOTA methods in eight evaluation metrics on two standard datasets (EORSSD and ORSSD). We also report the generalization performance of the model on the challenging ORSI-4199 to evaluate its robustness.

Diabetic retinopathy (DR), a serious complication of diabetes, is one of the primary causes of vision loss in retinal vascular diseases. While deep learning has been widely used for DR grading, their performance declines significantly when applied to data outside the training distribution due to domain shifts. Domain generalization (DG) addresses this challenge, yet most existing DG methods neglect lesion-specific features, limiting diagnostic accuracy.In this paper, we propose a novel approach that enhances existing DG methods by incorporating structural priors, inspired by the observation that DR grading is heavily dependent on vessel and lesion structures. We introduce Low-rank Adaptive Structural Priors (LoASP), a plug-and-play framework designed for seamless integration with existing DG models. LoASP improves generalization by learning adaptive structural representations that are finely tuned to the complexities of DR diagnosis.Extensive experiments on eight diverse datasets validate its effectiveness in both single-source and multisource domain scenarios. Visualizations show the learned priors align with vessel/lesion structures, enhancing interpretability and diagnostic relevance.

Contextual reasoning with constraints is crucial for enhancing temporal consistency in cross-frame modeling for visual tracking. However, mainstream tracking algorithms typically associate context by merely stacking historical information without explicitly supervising the association process, making it difficult to effectively model the target's evolving dynamics. To alleviate this problem, we propose RSTrack, which explicitly models and supervises context reasoning via three core mechanisms. 1) Context Reasoning Mechanism : Constructs a target state reasoning pipeline, converting unconstrained contextual associations into a temporal reasoning process that predicts the current representation based on historical target states, thereby enhancing temporal consistency. 2) Forward Supervision Strategy : Utilizes true target features as anchors to constrain the reasoning pipeline, guiding the predicted output toward the true target distribution and suppressing drift in the context reasoning process. 3) Efficient State Modeling : Employs a compression-reconstruction mechanism to extract the core features of the target, removing redundant information across frames and preventing ineffective contextual associations. These three mechanisms collaborate to effectively alleviate the issue of contextual association divergence in traditional temporal modeling. Experimental results show that RSTrack achieves state-of-the-art performance on multiple benchmark datasets while maintaining real-time running speeds. Our code is available at https://github.com/GXNU-ZhongLab/RSTrack.

The rich motion and appearance cues between consecutive frames are crucial for robust visual tracking. However, most existing tracking methods are still limited in designing different components to separately employ corresponding cues and even ignore one of them. This makes them difficult to maintain effective interaction between different cues, thus hindering the models from fostering a comprehensive understanding of the target objects. To address these issues, we propose a unified spatio-temporal cues learning framework (named USCLTrack) that comprehensively mines the variation patterns of targets between consecutive frames in complex video streams. Specifically, USCLTrack firstly aggregates motion and appearance cues into shared queries to provide the bridge of interaction between both cues. Then, it directly generates object locations on the condition of these shared queries in an autoregressive manner, unifying different cues to guide future inferences. To effectively learn multiple spatio-temporal cues aggregated in the shared queries, we develop a spatio-temporal attention mechanism. This mechanism integrates motion cues with appearance cues according to the time steps for ensuring temporal consistency. Moreover, it concurrently captures motion trends and appearance changes to facilitate the understanding of the target objects. Extensive experiments on eight popular tracking benchmarks validate the effectiveness of the proposed USCLTrack.

Abstract Facial emotion recognition in the wild is an important task in computer vision, but it still remains challenging since the influence of backgrounds, occlusions and illumination variations in facial images, as well as the ambiguity of expressions. This paper proposes a harmonious mutual learning framework for emotion recognition, mainly through utilizing attention mechanisms and probability distributions without utilizing additional information. Specifically, this paper builds an architecture with two emotion recognition networks and makes progressive cooperation and interaction between them. We first integrate self-mutual attention module into the backbone to learn discriminative features against the influence from emotion-irrelevant facial information. In this process, we deploy spatial attention module and convolutional block attention module for the two networks respectively, guiding to enhanced and supplementary learning of attention. Further, in the classification head, we propose to learn the latent ground-truth emotion probability distributions using softmax function with temperature to characterize the expression ambiguity. On this basis, a probability distribution distillation learning module is constructed to perform class semantic interaction using bi-directional KL loss, allowing mutual calibration for the two networks. Experimental results on three public datasets show the superiority of the proposed method compared to state-of-the-art ones.

Precise polyp segmentation is extremely challenging due to the diverse appearances and shaps’s of polyps along with severe light imbalance. To this end, we propose a novel network architecture, GroupSeg, which aims to learn robust representations by fully exploring global contextual representations and texture features of polyps through grouping segmentation to improve polyp segmentation performance. Specifically, GroupSeg groups features from the transformer encoder into progressively larger segments of different shapes, making full use of global contextual information to generate segmentation maps from fine to coarse, and thus can automatically adapt to polyps of different shapes and sizes, and is highly adaptive to the distribution of different polyp datasets. To further mine texture features to improve the performance of polyp segmentation, we introduce a Grouping Feature Aggregation Module (GFA), which adaptively mines the clues of local pixels and grouping feature segments, making the grouping features more accurate.GroupSeg demonstrates superior performance compared to state-of-the-art methods on ETIS, Endoscene, and ColonDB datasets, while high competitiveness on ClinicDB. It offers propsective outcomes in polyp segmentation.

To better restore a clean image from a noise observation under high noise levels, the authors propose an image denoising network based on the combination of multi‐scale and residual learning. Instead of using filters with different large sizes in traditional multi‐scale schemes, they arrange multi‐layer convolutions with the filters of the same size to speed up the model. Some dilated convolutions of different rates are combined with the common convolutions to enrich the extracted features in multi‐layer convolutions. Furthermore, they cascade the multi‐layer convolutions with residual blocks to improve the performance of image denoising. Their extensive evaluations on several challenging datasets demonstrate that the proposed model outperforms the state‐of‐art methods under all different noise levels in terms of peak signal‐to‐noise ratio, and the visual effects achieved by the proposed model are also better than the competing methods.

This letter proposes an ensemble neural network (Ensem-NN) for skeleton-based action recognition. The Ensem-NN is introduced based on the idea of ensemble learning, “two heads are better than one.” According to the property of skeleton sequences, we design one-dimensional convolution neural network with residual structure asBase-Net. From entirety to local, from focus to motion, we designed four different subnets based on theBase-Netto extract diverse features. The first subnet is aTwo-stream Entirety Net, which performs on the entirety skeleton and explores both temporal and spatial features. The second is aBody-part Net, which can extract fine-grained spatial and temporal features. The third is anAttention Net, in which a channel-wised attention mechanism can learn important frames and feature channels.Frame-difference Net, as the fourth subnet, aims at exploring motion features. Finally, the four subnets are fused as one ensemble network. Experimental results show that the proposed Ensem-NN performs better than state-of-the-art methods on three widely used datasets.

Robust vessel segmentation of fundus images is of great interest for better diagnosis of many diseases like diabetic retinopathy, retinopathy of prematurity, vein occlusions and so on. In this paper, we propose a novel example-based vessel segmentation method, based on learning the mapping relationship between fundus images and their corresponding ground truths. Firstly, the training images and their corresponding ground truths are divided into patches and clustered. Secondly, the mapping functions for each cluster are computed in a simple and efficient way from the training patches to their manual segmentation patches. Finally, Vessel segmentation are reconstructed by the simple mapping functions. Experimental results show that our method is efficient and can achieve competitive performance for vessel segmentation problems.

With the increasing demands for vehicle-to-vehicle and vehicle-to-infrastructure communications in intelligent transportation systems, new generation of vehicular telematics inevitably depends on the cooperation of heterogeneous wireless networks. In heterogeneous vehicular telematics, the network selection is an important step to the realization of multimode communications that use multiple access technologies and multiple radios in a collaborative manner. This paper presents an innovative network selection solution for the fundamental technological requirement of multimode communications in heterogeneous vehicular telematics. To guarantee the QoS satisfaction of multiple mobile users and the efficient utilization and fair allocation of heterogeneous network resources in a global sense, a dynamic and self-adaptive method for network selection is proposed. It is biologically inspired by the cellular gene network, which enables terminals to dynamically select an appropriate access network according to the variety of QoS requirements and to the dynamic conditions of various available networks. The experimental results prove the effectiveness of the bioinspired scheme and confirm that the proposed network selection method provides better global performance when compared with the utility function method with greedy optimization.