Recent advancements in deep learning-based detection methods have significantly improved infrared small target detection (IRSTD). However, these approaches often require costly pixel-level annotations to achieve satisfactory performance. To address this challenge, we propose a weakly semi-supervised infrared small target detection (WSSISTD) framework. This framework comprises a teacher network that generates coarse pseudo-masks and a student network that focuses on the final detection task. In WSSISTD, we initially utilize a minimal proportion of pixel-level annotated data to train the teacher network, which subsequently generates coarse pseudo-masks for a substantial volume of cost-effective point annotation data. In addition, an adaptive random walker (RW) strategy is introduced to fully leverage point labels for generating RW pseudo-masks. Furthermore, to mitigate the limitations of the teacher network, which is trained exclusively on a limited amount of pixel-level labels, we have developed a RefineNet to integrate and refine multiple coarse pseudo-masks. Ultimately, a combination of partially real masks and numerous refined pseudo-masks is employed to jointly train the student network, thereby effectively reducing the labeling budget. To validate the effectiveness of this method, extensive experiments were conducted on three datasets. The results indicate that, with pixel-level labels constituting merely 10% of the data, our method achieves performance comparable with fully supervised methods in segmentation (intersection over union (IoU): 70.31% versus 73.36%) and detection (probability of target detection (Pd): 93.58% versus 93.67%).

The popularity of immersive videos has prompted extensive research into neural adaptive tile-based streaming to optimize video transmission over networks with limited bandwidth. However, the diversity of users’ viewing patterns and Quality of Experience (QoE) preferences has not been fully addressed yet by existing neural adaptive approaches for viewport prediction and bitrate selection. Their performance can significantly deteriorate when users’ actual viewing patterns and QoE preferences differ considerably from those observed during the training phase, resulting in poor generalization. In this paper, we proposeMANSY, a novel streaming system that embraces user diversity to improve generalization. Specifically, to accommodate users’ diverse viewing patterns, we design a Transformer-based viewport prediction model with an efficient multi-viewport trajectory input output architecture based on implicit ensemble learning. Besides, we for the first time combine the advanced representation learning and deep reinforcement learning to train the bitrate selection model to maximize diverse QoE objectives, enabling the model to generalize across users with diverse preferences. Extensive experiments demonstrate thatMANSYoutperforms state-of-the-art approaches in viewport prediction accuracy and QoE improvement on both trained and unseen viewing patterns and QoE preferences, achieving better generalization.

Semantic segmentation in nighttime scenes presents a significant challenge in autonomous driving. Unsupervised domain adaptation (UDA) offers an effective solution by learning domain-invariant features to transfer models from the source domain (daytime scenes) to the target domain (nighttime scenes). Many methods introduce a latent domain to reduce the difficulty of UDA. However, they often build only a single adaptation pair of “latent-to-target”, which limits the effectiveness of knowledge transfer across different domains. In this letter, we propose a Masked Guided Meta-Learning (MGML) framework for domain-adaptive nighttime semantic segmentation. Within the MGML framework, we explore two key issues: how to generate the latent domain, and how to leverage the latent domain to assist meta-learning in reducing domain discrepancy. For the first issue, we employ the fast Fourier transform along with a complementary masking strategy to generate masked latent images that resemble the target scenes in the latent domain without adding to the training burden. For the second issue, we nest a mask-based consistency constraint within a bi-level meta-learning framework, enabling cross-domain knowledge acquired from the pair of “source-to-latent” to enhance the “latent-to-target” adaptation. Experiments on benchmark datasets demonstrate that our MGML achieves state-of-the-art performance, demonstrating the effectiveness of our approach in nighttime semantic segmentation.

In this letter, a new reconstructed variational Bayesian (RVB) Kalman filter is proposed for handling heavy-tailed and skewed noises. The Student's t distribution is used to model predicted state, and the skewed Gaussian-exponential mixture distribution is used to model measurement noises. In order to more accurately deal with process noises, we model the predicted covariance based on the inverse Wishart distribution. Subsequently, a new RVB framework is proposed to separate the iterations of process and measurement noise related parameters, which achieves decoupling of hidden variables. Under the new RVB framework, the RVB-ST-SGEM-KF algorithm is derived. The superiority of the proposed RVB framework and algorithm are validated through a maneuvering target tracking scenario.

One of the most important features of next-generation communication is to incorporate intelligence towards semantic communication, where highly condensed semantic information considering both source and channel features will be extracted and transmitted. The recent popular large models such as GPT4 and the boosting learning techniques are envisioned to accelerate its practical implementation in the near future. Given the characteristics of “training once and widely use” of those multimodal large language models, we argue that a pay-as-you-go service mode will be suitable in this context, referred to as Large Model as a Service (LMaaS). However, the trading and pricing problem is quite complex with heterogeneous and dynamic customer environments, making the pricing optimization problem challenging in seeking on-hand solutions. In this paper, we optimize the profit of both the seller and customers. We formulate the LMaaS market trading as a Stackelberg game with two steps. In the first step, we optimize the seller's pricing decision and propose an Iterative Model Pricing (IMP) algorithm that optimizes the prices of large models iteratively by reasoning customers’ future rental decisions, which achieves a near-optimal pricing solution. In the second step, we optimize customers’ selection decisions by designing a robust selecting and renting (RSR) algorithm, which is guaranteed to be optimal with rigorous theoretical proof. Extensive experiments confirm the effectiveness and robustness of our algorithms, outperforming the state-of-the-art solution by 43.96% in profit at the customer side and achieving the near-optimal profit at the seller side.

Foundation models (FMs) have shown great success in natural language processing, computer vision, and multimodal tasks. FMs have a large number of model parameters, thus requiring a substantial amount of data to help optimize the model during the training. Federated learning has revolutionized machine learning by enabling collaborative learning from decentralized data while still preserving clients’ data privacy. Despite the great benefits foundation models can have empowered by federated learning, their bulky model parameters cause severe communication challenges for modern networks and computation challenges especially for edge devices. Moreover, the data distribution of different clients can be different thus inducing statistical challenges. In this paper, we propose a novel two-stage federated learning algorithm called FedFMSL. A global expert is trained in the first stage and a local expert is trained in the second stage to provide better personalization. We construct a Mixture of Foundation Models (MoFM) with these two experts and design a gate neural network with an inserted gate adapter that joins the aggregation every communication round in the second stage. To further adapt to edge computing scenarios with limited computational resources, we design a novel Sparsely Activated LoRA (SAL) algorithm that freezes the pre-trained foundation model parameters inserts low-rank adaptation matrices into transformer blocks, and activates them progressively during the training. We employ extensive experiments to verify the effectiveness of FedFMSL, results show that FedFMSL outperforms other SOTA baselines by up to 59.19% in default settings while tuning less than 0.3% parameters of the foundation model.

A novel near-space floating platform aided passive location method for super-low-altitude target using time difference of arrival(TDOA) is proposed. The new method uses a near-space floating platform as the baseline in the height to form a redundant location system consisting of several subsystems. Each subsystem can obtain a set of location results with ambiguity. After eliminating the ambiguity by the nearest matching, the new method can improve the location accuracy in the whole observed airspace by simplified weighted least square(SWLS) fusion. A better tracking performance can be achieved by treating the SWLS fusion location result as pseudo-linear measurement, and keeping the tracking through Kalman filter. Simulation results show that the proposed method comes to stabilization within 40s and has higher tracking precision than the SWLS method.