Deep learning models often suffer from performance degradation in unseen domains, posing a risk for safety-critical applications such as autonomous driving. To tackle this problem, recent studies have leveraged pre-trained Visual Foundation Models (VFMs) to enhance generalization. However, exsiting works mainly focus on designing intricate networks for VFMs, neglecting their inherent strong generalization potential. Moreover, these methods typically perform inference on low-resolution images. The loss of detail hinders accurate predictions in unseen domains, especially for small objects. In this paper, we argue that simply fine-tuning VFMs and leveraging high-resolution images unleash the power of VFMs for generalizable semantic segmentation. Therefore, we design a VFM-based segmentation network (VFMNet) that adapts VFMs to this task with minimal fine-tuning, preserving their generalizable knowledge. Then, to fully utilize high-resolution images, we train a Mask-guided Refinement Network (MGRNet) to refine VFMNet's predictions combining detailed image features. Furthermore, we adopt a two-stage coarse-to-fine inference approach. MGRNet is used to refine the low-confidence regions predicted by VFMNet to obtain fine-grained results. Extensive experiments demonstrate the effectiveness of our method, outperforming state-of-the-art methods by 3.3% on the average mIoU in synthetic-to-real domain generalization.

MmWave imaging assisted by metasurfaces is a burgeoning technique attributed to its fine-grained imaging ability by time-varying phase coding, which produces a large virtual aperture. To obtain stereoscopic 3D perception, multiple metasurfaces can be utilized for virtual multiview fusion, allowing for the capture of features that are not visible from a single viewpoint. However, the multipath imaging fusion faces huge data burden and the environment clutter especially reflection from the direct path will cause disturbance. To address these issues, this paper introduces Bayesian compressive sensing to focus on the region of interest (ROI) and design a double sparse prior for high-resolution multiview image reconstruction. First, multiple metasurfaces are utilized to generate virtual multiview imaging results. Then, the Bayesian inference method is leveraged to resist environmental noise and achieve autofocusing imaging with undersampled data. The expectation propagation (EP) is introduced to estimate the statistical parameter iteratively. Further, a double sparse prior is designed based on spike-and-slab to promote inter-sparsity and intra-sparsity simultaneously. Simulation results show that our proposed system demonstrates superior performance by fusing sparse images from multiple metasurfaces arrangements.

Collecting and annotating data for semantic segmentation can end up costing a lot of time and energy. Unsupervised Domain Adaptation (UDA) for semantic segmentation allows models trained on certain source domain data (such as the GTA synthetic dataset) to be applied to certain target data (like the Cityscapes real dataset), significantly decreasing the requirement for target-domain manual pixel-level annotations.In this paper, we suggest a cross-domain unified semantic segmentation network training framework, the Attention Distribution Adaptation Network (ADAN). It (1) proposes the Content Consistency Transformation (CCT), which maps source domain data to an intermediate domain that has a data distribution that is comparable to that of the target domain, and (2) introduces the Attention Adaptation Module (AAM), which bolsters the synchronization of attention between the intermediate domain and the target domain. ADAN achieved unprecedented mIoUs of 77.2 and 68.8 on GTA→Cityscapes and Synthia→Cityscapes, respectively, corresponding to improvements of +1.3 and +0.6 over the state of the art.

A common approach for accelerating optimization algorithms is to minimize the loss achieved in a fixed time, which enables a differentiable framework with respect to the algorithm's hyperparameters. In contrast, the complementary objective of minimizing the time to reach a target loss is traditionally considered non-differentiable. To address this limitation, we propose a differentiable discrete stopping time and theoretically justify it based on its connection to continuous differential equations. We design an efficient algorithm to compute its sensitivities, thereby enabling a new differentiable formulation for directly accelerating algorithms. We demonstrate its effectiveness in applications such as online hyperparameter tuning and learning to optimize. Our proposed methods show superior performance in comprehensive experiments across various problems, which confirms their effectiveness.

Data deduplication technique has been widely adopted by commercial cloud storage providers, which is both important and necessary in coping with the explosive growth of data. To further protect the security of users’ sensitive data in the outsourced storage mode, many secure data deduplication schemes have been designed and applied in various scenarios. Among these schemes, secure and efficient re-encryption for encrypted data deduplication attracted the attention of many scholars, and many solutions have been designed to support dynamic ownership management. In this paper, we focus on the re-encryption deduplication storage system and show that the recently designed lightweight rekeying-aware encrypted deduplication scheme (REED) is vulnerable to an attack which we call it stub-reserved attack. Furthermore, we propose a secure data deduplication scheme with efficient re-encryption based on the convergent all-or-nothing transform (CAONT) and randomly sampled bits from the Bloom filter. Due to the intrinsic property of one-way hash function, our scheme can resist the stub-reserved attack and guarantee the data privacy of data owners’ sensitive data. Moreover, instead of re-encrypting the entire package, data owners are only required to re-encrypt a small part of it through the CAONT, thereby effectively reducing the computation overhead of the system. Finally, security analysis and experimental results show that our scheme is secure and efficient in re-encryption.

Proof of replication technique has been widely used to verify whether the cloud service providers (CSPs) store multiple replications of a file with dedicated and unique storage space, which effectively prevents CSPs from colluding and storing only one copy of the file. In this field, many representative schemes have been proposed and applied to various scenarios. However, most of the existing schemes are based on the timing assumption (i.e., the verifier rejects the proof of replication if the prover's response is timeout) and do not explicitly consider the problem of batch verification and fault localization. This will bring unnecessary computational overhead to the verifier and reduce the efficiency of batch auditing. To address the above problems, we propose a verifiable proof of replication scheme with fast fault localization and high efficiency. By integrating incompressible encoding and homomorphic linear authenticator, our scheme can effectively audit the integrity of file replications without timing assumptions. To support batch verification and fault localization, we propose a reversed signature aggregation tree (Rev-tree) by integrating the quick binary search and exponent testing. Compared with the traditional binary tree, Rev-tree can further reduce the overhead of batch verification and effectively locate a single fault replication. Moreover, benefit from the property of Rev-tree taking the existing error probability as an estimate of the rest of the tree, our scheme can adjust the verification strategy dynamically to meet with different situations. Finally, security analysis and experimental results show that our scheme is secure and efficient in proof of replication and fast fault localization.