For pursuing accurate skeleton-based action recognition, many existing graph-based approaches deploy the higher-order polynomials of the skeletal adjacency matrix to model the node correlations of distant neighbours. To further capture robust graphical patterns, a novel multi-scale graph convolution operator is proposed, which enables to aggregate multi-scale dependencies and capture long-range joint relationships on human skeleton graph. Additionally, a novel corrective contrastive learning strategy is proposed, which aims to distinguish the representative clues and calibrate the confused action clips in the feature space. Comprehensive experiments validate the effectiveness and superiority of our proposed method over state-of-the-art approaches on three large-scale datasets: NTU RGB+D 60, NTU RGB+D 120, and Kinetics Skeleton 400.

In AIoT-enabled secure and green supply chain systems, robust device authentication measures are crucial to maintaining the integrity of the ecosystem. One key challenge in this context is mitigating password guessing attacks—a scenario that can be modeled as a specialized sequence prediction task demanding high character-level accuracy and computational efficiency to safeguard devices and data. Although natural language processing (NLP) models, particularly GPT-based approaches, have excelled in sequence prediction tasks, they often lack the precision needed for password guessing due to their reliance on broad contextual dependencies. To address these limitations, we propose PassGAT, a novel framework that leverages Graph Attention Networks (GAT) to enhance password prediction performance. PassGAT transforms passwords into graph representations, where each character is treated as a node, enabling selective computation of attention coefficients between characters. This approach captures both local dependencies essential for character-level accuracy and global patterns that enhance the understanding of password structures. Experimental results demonstrate that PassGAT achieves an average improvement of 15.58% in accuracy over an existing GPT-based password guessing model while reducing computational overhead by 85.19%. By significantly enhancing authentication accuracy and efficiency, PassGAT provides a robust and sustainable solution to mitigate password-related security risks in AIoT-enabled supply chain systems.

Due to the high cost of pixel-level labels required for fully-supervised semantic segmentation, weakly-supervised segmentation has emerged as a more viable option recently. Existing weakly-supervised methods tried to generate pseudo-labels without pixel-level labels for semantic segmentation, but a common problem is that the generated pseudo-labels contain insufficient semantic information, resulting in poor accuracy. To address this challenge, a novel method is proposed, which generates class activation/attention maps (CAMs) containing sufficient semantic information as pseudo-labels for the semantic segmentation training without pixel-level labels. In this method, the attention-transfer module is designed to preserve salient regions on CAMs while avoiding the suppression of inconspicuous regions of the targets, which results in the generation of pseudo-labels with sufficient semantic information. A pixel relevance focused-unfocused module has also been developed for better integrating contextual information, with both attention mechanisms employed to extract focused relevant pixels and multi-scale atrous convolution employed to expand receptive field for establishing distant pixel connections. The proposed method has been experimentally demonstrated to achieve competitive performance in weakly-supervised segmentation, and even outperforms many saliency-joined methods.

Medical images often contain sensitive information, and one typical security measure is to encrypt medical images prior to storage and analysis. A number of solutions, such as those utilizing deep learning, have been proposed for medical image encryption and decryption. However, our research shows that deep learning-based encryption models can potentially be vulnerable to backdoor attacks. In this paper, a backdoor attack paradigm for encryption and decryption network is proposed and corresponding attacks are respectively designed for encryption and decryption scenarios. For attacking the encryption model, a backdoor discriminator is adopted, which is randomly trained with the normal discriminator to confuse the encryption process. In the decryption scenario, a number of subnetwork parameters are replaced and the subnetwork can be activated when detecting the trigger embedded into the input (encrypted image) to degrade the decryption performance. Considering the model performance degradation due to parameter replacement, the model pruning is also adopted to further strengthen the attacking performance. Furthermore, the image steganography is adopted to generate invisible triggers for each image; subsequently, improving the stealthiness of backdoor attacks. Our research on designing backdoor attacks for encryption and decryption network can serve as an attacking mode for such networks, and provides another research direction for improving the security of such models. This research is also one of the earliest works to realize the backdoor attack on the deep learning based medical encryption and decryption network to evaluate the security performance of these networks. Extensive experimental results show that the proposed method can effectively threaten the security performance both for the encryption and decryption network.