The revolution of information technologies is blur-ring the boundary between physical and virtual worlds, with digital twin (DT) at the forefront of this transformation. Mean-while, DT is evolving from offering simple data mirrors to smart copilots, capable of understanding user’s intents and proactively providing profound insights. This ushers in the pivotal cognitive stage of the DT, where humans or embodied artificial intelligence agents (EAIs) can obtain cognitive perceptions within the DT world. Among various issues, in the cognitive stage, establishing strong interactions between the physical and DT worlds is critical. This paper thus explores the Next-Generation Semantic Communication (NextG-SemCom), well-suited for this issue, to fully enable the DT world with human-like cognitive capabilities. NextG-SemCom is envisioned as a cognitive-native paradigm that leverages Large AI Models (LAMs) as its core engine to perform intent understanding, contextual planning, and intent-oriented generative extraction. NextG-SemCom establishes a closed cognitive loop that shifts communication from data-driven to cognitive-driven, thereby not only reducing the traffic overhead, but also achieving cognitive comprehension of the transmitted information. This paper provides the first survey of the NextG-SemCom driven DT world, focusing on the advancement, trend and vision. We start by tracing the evolution of the DT from the initial virtual mapping stage to the current cognitive stage, and detail the distinguishing features of NextG-SemCom. We then present a holistic framework of cognitive interactions in the DT world, analyzing its core components, including the NextG-SemCom codec and network management, along with its design requirements and challenges. Furthermore, we discuss potential applications across a spectrum of human-human, human-EAI, and EAI-EAI interactive scenarios. Finally, we outline future research directions to provide a roadmap and inspire further studies in this promising field.

Local feature descriptor learning aims to represent distinctive images or patches with the same local features, where their representation is invariant under different types of deformation. Recent studies have demonstrated that descriptor learning based on Convolutional Neural Network (CNN) is able to improve the matching performance significantly. However, they tend to ignore the importance of sample selection during the training process, leading to unstable quality of descriptors and learning efficiency. In this paper, a dual hard batch construction method is proposed to sample the hard matching and non-matching examples for training, improving the performance of the descriptor learning on different tasks. To construct the dual hard training batches, the matching examples with the minimum similarity are selected as the hard positive pairs. For each positive pair, the most similar non-matching example is then sampled from the generated hard positive pairs in the same batch as the corresponding negative. By sampling the hard positive pairs and the corresponding hard negatives, the hard batches are produced to force the CNN model to learn the descriptors with more efforts. In addition, based on the above dual hard batch construction, an ℓ22 triplet loss function is built for optimizing the training model. Specifically, we analyze the superiority of the ℓ22 loss function when dealing with hard examples, and also demonstrate it in the experiments. With the benefits of the proposed sampling strategy and the ℓ22 triplet loss function, our method achieves better performance compared to state-of-the-art on the reference benchmarks for different matching tasks.

In recent years, multitrack music generation has garnered significant attention in both academic and industrial spheres for its versatile utilization of various instruments in collaborative settings. The primary challenge lies in achieving a harmonious balance within individual tracks and fostering effective collaboration across multiple tracks. To address this issue, this article introduces a pioneering hybrid learning encoder architecture. Each music track's encoder is implemented as an independent transformer architecture, preserving self-attention mechanisms within a single track and interattention mechanisms between different tracks. The resulting features are then seamlessly integrated into the decoder through concatenation. Of particular significance, previous multitrack music generation efforts have predominantly operated under unconditional settings, yielding music that lacks practical value due to noncompliance with established music theory principles. Recognizing this limitation, the article proposes a novel approach to multitrack music generation guided by music theory rules. Employing reinforcement learning techniques, the decoder-generated music serves as the initial state. Positive feedback is provided when the generated music adheres to music theory rules; conversely, negative feedback is applied to compel the multitrack music to align with widely accepted music theory principles. Finally, comprehensive simulation validation is conducted on both the publicly available LMD dataset and the self-constructed MUT dataset. The plethora of experimental results overwhelmingly corroborates the efficacy of the proposed methodology.

Region-of-interest (ROI) detection plays a significant role in the analysis and interpretation of remote sensing images (RSI), due to the huge size of satellite images and their explosive growth in quantity. However, when applied to panchromatic RSI directly, traditional saliency models cannot achieve satisfying performance for two reasons: one is the computational efficiency decrease caused by the huge image size; the other is the absence of color information for panchromatic RSI. Thus, in this letter, an ROI detection model based on statistical distinctiveness (SD) is proposed for saliency analysis and ROIs detection in panchromatic RSI. The proposed SD model incorporates both the lower order SD (LSD) and the higher order SD (HSD), in order to identify regions of interest that are highly distinctive from the rest of the scene. Finally, the saliency map is determined by fusing cue maps obtained by calculating LSD locally and HSD globally. Experimental results show that our approach achieves promising results when compared with existing state-of-the-art saliency detection models.