Network traffic classification is crucial for cybersecurity, as it helps detect abnormal traffic and potential attack patterns, enabling timely protective measures. With the widespread adoption of remote work, video conferencing, and VPN services, the rapid growth of network traffic has increased the complexity of traffic classification. Traditional methods, such as port-based classification, Deep Packet Inspection, and Machine Learning (ML), face limitations in handling complex and encrypted traffic. In this study, we propose a novel network traffic classification system that combines 1D-CNN and Transformer, utilizing different activation functions to extract both local and long-range features. These complementary features are fused together to improve classification capability. The experimental results on the ISCX VPN-nonVPN dataset indicate that our proposed system outperforms baseline methods in most categories with respect to Precision, Recall, and F1-score.

Urbanization, characterized by the continuous growth of population and density in urban areas, has led to the expansion and increased complexity of transportation networks, exacerbating traffic congestion. Accurate traffic speed prediction is crucial for effective traffic network management and planning. As the complexity of real-world road networks increases, integrating spatial and temporal information for accurate traffic speed prediction has become a challenging research task. This paper proposes a novel approach by introducing a spatial-temporal graph neural network (STGNN)-based model to enhance the accuracy of traffic speed prediction. By employing an attention-based STGNN, we effectively capture the complex relationships among road segments in real-world scenarios. We utilize the Huber loss as the training objective to improve prediction accuracy. Furthermore, we present an architecture that incorporates Root Mean Square Layer Normalization into the Transformer and integrates the Spatial-Temporal Attention Wavenet (STAWnet) into the model backbone, enabling faster training while maintaining model stability. We evaluate the proposed model using five real-world traffic speed benchmark datasets. The experimental results demonstrate that our method achieves superior performance compared to state-of-the-art traffic speed prediction approaches.

Cyber Threat Intelligence (CTI) significantly enhances organizational cybersecurity defenses by providing actionable insights from diverse data sources. This research studies the correlation between CTI analysis and the MITRE ATT&CK framework, focusing on their alignment to strengthen threat detection and response capabilities. A pivotal aspect of this study involves developing a classifier using a fine-tuned BERT-based model to map CTI reports to specific ATT&CK techniques. Our model demonstrated substantial improvements over the baseline SecBERT, achieving a 2.6% higher F1-score and a 4.2% improvement in Top-3 Accuracy. By integrating CTI with the MITRE ATT&CK framework, researchers can shift from reactive to proactive cybersecurity strategies. This integration enables swift detection of emerging threats, enhances incident response effectiveness, and fortifies defensive measures against evolving cyber threats. Ultimately, the synergy between CTI and ATT&CK fosters a comprehensive approach to cybersecurity management in today’s dynamic threat landscape.

This paper pushes the envelope on decomposing camouflaged regions in an image into meaningful components, namely, camouflaged instances. To promote the new task of camouflaged instance segmentation of in-the-wild images, we introduce a dataset, dubbed CAMO++, that extends our preliminary CAMO dataset (camouflaged object segmentation) in terms of quantity and diversity. The new dataset substantially increases the number of images with hierarchical pixel-wise ground truths. We also provide a benchmark suite for the task of camouflaged instance segmentation. In particular, we present an extensive evaluation of state-of-the-art instance segmentation methods on our newly constructed CAMO++ dataset in various scenarios. We also present a camouflage fusion learning (CFL) framework for camouflaged instance segmentation to further improve the performance of state-of-the-art methods. The dataset, model, evaluation suite, and benchmark will be made publicly available on our project page.

Estimating the mask-wearing ratio in public places is important as it enables health authorities to promptly analyze and implement policies. Methods for estimating the mask-wearing ratio on the basis of image analysis have been reported. However, there is still a lack of comprehensive research on both methodologies and datasets. Most recent reports straightforwardly propose estimating the ratio by applying conventional object detection and classification methods. It is feasible to use regression-based approaches to estimate the number of people wearing masks, especially for congested scenes with tiny and occluded faces, but this has not been well studied. A large-scale and well-annotated dataset is still in demand. In this paper, we proposed two different methods for ratio estimation that are leveraged by either detection-based or regression-based approaches. For the detection-based approach, we improved a state-of-the-art face detector, RetinaFace, for the ratio estimation. For the regression-based approach, we utilized a baseline network, CSRNet, and finetuned it to estimate the density maps for masked and unmasked faces. We also proposed the first large-scale dataset, the “NFM,” which contains 581,108 face annotations extracted from 18,088 video frames in 17 street-view videos11The annotations (bounding boxes and labels), and pre-trained models will be released with the publication of our paper. , Through experiments, the RetinaFace-based method achieves better accuracy under different situations, while the CSRNet-based method is superior in terms of operation time thanks to its compactness.

Salient object detection aims to detect the main objects in the given image. In this paper, we proposed an approach that integrates semantic priors into the salient object detection process. The method first obtains an explicit saliency map that is refined by the explicit semantic priors learned from data. Then an implicit saliency map is constructed using a trained model that maps the implicit semantic priors embedded into superpixel features with the saliency values. Next, the fusion saliency map is computed by adaptively fusing both the explicit and implicit semantic maps. The final saliency map is eventually computed via the post-processing refinement step. Experimental results have demonstrated the effectiveness of the proposed method, particularly, it achieves competitive performance with the state-of-the-art baselines on three challenging datasets, namely, ECSSD, HKUIS, and iCoSeg.