Recent weakly supervised image dehazing (WSID) works have succeeded to improve models’ generalization ability to real scene dehazing by using generative adversarial network (GAN) for unpaired image training. However, it is still difficult for current WSID methods to train one effective dehazing model for various scenes since 1) they always result in residual haze due to insufficient generalization to the feature distribution of real scenes, and 2) they are prone to cause distortions like color shifts, artifacts or halos etc, owing to embedding manual prior or threshold hypothesis for image reconstruction. To solve above problems, in this paper, we propose a novel WSID model via physics-based decomposition (PBD), which estimates atmospheric light, scattering coefficient and scene depth of real haze input to effectively capture the illumination information and haze distribution to recover a preliminary dehazed image by minimizing reconstruction loss. With this constraint, we subtly design a discrete wavelet discriminator (DWD) to effectively improve the generalization to real scene from both spatial and frequency aspect under the supervision of unpaired real clear image. Our PBD is a purely data-driven model freeing from any manual setting or partially correct prior, thus simultaneously ensuring the realness and visibility of dehazed images. Experiments on seven benchmarks verified the strong generalization ability of our PBD, which achieves SOTA dehazing performance with realistic details. Code will be published at https://github.com/NianWang-HJJGCDX/PBD.

Collaborative camouflaged object segmentation (CoCOS) is a challenging task, focusing on identifying objects that blend closely with their backgrounds by jointly processing intraclass images. Existing methods fail to fully leverage the shared features (e.g., shape, texture, and contour) from these intraclass images, which leads to poor segmentation performance in relatively complex scenarios. To address this issue, we propose a novel mutually guided fusion refinement network (MFRNet), which improves the model performance by more effectively collaborating and optimizing the shared information. Specifically, it includes feature encoding, single-image branch feature enhancement, multiimage branch feature enhancement, and mutual guidance. After the feature encoding step, we design the graph convolution self-attention (GCS) and spatial context exploration (SCE) modules to enhance multilevel features of the single-image and multiimage branches, respectively. Moreover, we propose a mutual guidance fusion (MGF) module to utilize cross-scene image information for mutual guidance and progressive refinement, enhancing intraclass collaboration for improving target feature distinction. Extensive experimental results demonstrate that our MFRNet significantly outperforms existing CoCOS methods, achieving a mean E-measure score of 0.846 on the CoCOD8K dataset. Our code will be published at https://github.com/another-u/MFRNet.

Recent advancements in unpaired dehazing, particularly those using GANs, show promising performance in processing real-world hazy images. However, these methods tend to face limitations due to the generator's limited transport mapping capability, which hinders the full exploitation of their effectiveness in unpaired training paradigms. To address these challenges, we propose DehazeSB, a novel unpaired dehazing framework based on the Schrödinger Bridge. By leveraging optimal transport (OT) theory, DehazeSB directly bridges the distributions between hazy and clear images. This enables optimal transport mappings from hazy to clear images in fewer steps, thereby generating high-quality results. To ensure the consistency of structural information and details in the restored images, we introduce detail-preserving regularization, which enforces pixel-level alignment between hazy inputs and dehazed outputs. Furthermore, we propose a novel prompt learning to leverage pre-trained CLIP models in distinguishing hazy images and clear ones, by learning a haze-aware vision-language alignment. Extensive experiments on multiple real-world datasets demonstrate our method's superiority. Code: https://github.com/ywxjm/DehazeSB.

ABSTRACT To address the issue of recovery imbalance caused by the spatial heterogeneity of haze concentration in real scenarios, this paper proposes an adaptive feature enhanced contrastive learning framework (AFE‐Dehaze). The framework achieves breakthroughs through three major collaborative mechanisms: (1) a hierarchical multi‐scale fusion architecture that combines diffusion convolution and channel attention, preserving edge textures (such as leaf veins and building contours) in thin haze areas, while semantically guiding the reconstruction of structural details in dense haze areas, improving texture retention by 18% compared to traditional U‐Net; (2) a concentration‐sensitive contrastive learning paradigm that uses pre‐trained VGG features as semantic anchors, applying pixel‐level constraints in thin haze and feature space constraints in dense haze, which reduces color distortion () by 23%, significantly outperforming methods like refusion; (3) a gradient dynamic balancing strategy that automatically adjusts the optimization direction by analyzing positive and negative sample gradient contributions, enhancing PSNR by 1.2dB and SSIM by 0.05 in non‐uniform haze scenarios. Experiments on a mixed dataset (RESIDE OTS real scenes) demonstrate that AFE‐Dehaze achieves an average PSNR of 28.7dB and SSIM of 0.91, especially improving structural similarity in dense haze areas by 9% compared to Mamba, validating its generalization capability in complex haze environments. This framework provides a solution that balances accuracy and robustness for dehazing in real scenarios such as vehicular vision and remote sensing imaging.

Image dehazing is essential to boost the visual quality of images captured in hazy conditions. Recently, many learning-based methods were proposed to achieve single image dehazing with the training of tremendous paired synthetic hazy/ real clean images. Due to the domain gap between real and synthetic scenes, these models cannot generalize well to various real hazy scenes, leading to under-dehazed results. To overcome this problem, we propose a real scene image Dehazing Network with Multi-prior Guidance and Domain Transfer (DNMGDT). Our DNMGDT is based on a parameter shared architecture trained by synthetic hazy images and real hazy images simultaneously. For real hazy images, multiple prior-based dehazed images are adopted as pseudo clean images. An Image Quality Guided Adaptive Weighting (IQGAW) scheme is proposed to form the supervision by automatically weighting different parts of these prior-based dehazed images and suppressing negative information of them. Moreover, to reduce the domain gap between real and synthetic hazy scenes, a Physical Model Guided image level Domain Transfer (PMGDT) mechanism is proposed to regularize the learning process with consistency constraint. Experiments on various datasets demonstrated the effectiveness of our proposed method especially for real hazy scenes.

In this paper, the dynamic modeling and collaborative control of a multi-vehicle cooperative transportation system for load carrying is explored. A hierarchical modeling and constraint-following control scheme is creatively proposed. In the dynamic modeling stage, the separate models of system components including the load and vehicle carriers are firstly established at the lower level. Then the internal and external constraints corresponding to the system topology and the transportation task are designed to integrate separate models at a higher level. In the system control stage, a distributed collaborative control law is proposed based on the closed-form constraint forces, with which the load can follow the external constraints actively and the carriers can maintain the internal constraints passively. In order to overcome the influence of time-varying multi-source uncertainties of the system on control effectiveness and stability, an adaptive robust control term is designed based on the Lyapunov min-max approach. Both uniform boundedness and uniform ultimate boundedness of the constraint-following error are guaranteed. Comprehensive validations show that our propose scheme can significantly reduce the modeling complexity despite the strongly coupled topology and nonlinearity of the system, as well as achieving more precise and robust trajectory following control compared with the baseline methods.

In this article, the trajectory generation and motion control of autonomous driving oversized cargo cooperative transportation systems (CTS) in static but bounded environment is investigated. Different from common vehicle systems, the challenges lie on the safety-guaranteed cooperation of independently controlled carriers with inherent connections brought by the rigid payload, which results in complex system dynamics and multiple time-variant uncertainties. A constraint-oriented “leader-follower” modeling and control framework is introduced, and a trajectory generation method based on the diffeomorphism is creatively proposed to generate closed-form collision-free trajectory for the payload in the bounded environment. To achieve safety-guaranteed trajectory following under uncertainties, a transformed adaptive robust control strategy (TARC) is designed through constraint relaxation, and the coordination of the carriers is realized. An implementation with comprehensive ablation studies demonstrates the effectiveness of our trajectory generation and tracking control framework. The collision-free trajectory set is efficiently generated, and the CTS can be kept strictly inside the safe corridor with high tracking accuracy, which is extremely hard for the baseline methods.

3D object detectors based on LiDAR have been extensively used in autonomous and robotic systems. Efficient voxel-based models must downsample their feature space to reduce computation, which leads to the loss of geometric information and limit their accuracy. To solve this problem, this paper presents a 3D detection framework, point-voxel and bird’s-eye-view representation aggregation network for single stage 3D object detection (PVB-SSD), in which a position information input branch generates Fourier embedding features from the origin point cloud to supplement the lost information. A global-former module integrates embedded Fourier features with bird’s-eye-view features extracted by a 3D convolution backbone. Considering that in the deeper layer of the neural network, the spatial level features will be replaced by semantic level features, a windows transformer spatial-semantic aggregate module fuses them dynamically. Extensive experiments on the KITTI, Waymo and NuScences datasets show that our model has excellent accuracy and relatively low computational consumption.

Future autonomous vehicles will incorporate high date rate communications and high-accuracy radar sensing capabilities operating in the millimeter-wave (mmWave) and higher frequencies, which results in Integrated sensing and communication (ISAC). Hybrid beamforming (HBF) is an attractive technology for practical vehicular ISAC systems. The HBF with the partially-connected structure (PCS) can effectively reduce the hardware cost and power consumption compared to fully-connected structure (FCS). But the constant-modulus constraint caused by PCS makes the HBF design problem non-convex, which poses a greater challenge. In this paper, we consider the HBF design with PCS as a weighted minimization problem of the communication and radar beamforming errors under the constant-modulus constraints and power constraints. Dual functions of communication and radar are expressed as a tradeoff in this question. Despite the optimization problem being non-convex and hard to obtain the global minimizer, we reduce the problem into a two-step subproblem including the analog precoder design and digital precoder design. Then, a variable projection-based alternating minimization algorithm is proposed to solve these problems. Unlike previous works, which focused on the relationship between variables to iteratively solve, our method exploits the intrinsic geometric features of the mmWave channel and the variable projection to simplify the solution of the beamformers. Simulation results demonstrate that the proposed algorithm achieves significantly improved performance in terms of the system spectral efficiency over the existing solutions and greatly reduces the computational complexity.

This letter proposes a multipath subspace projection andextreme learning machine(ELM)-based indoor fingerprint localization algorithm called MuSpel-Fi, where thechannel state information(CSI) is utilized as the raw data to establish fingerprints. In this algorithm, the CSI is firstly organized into a time-domain matrix and then is projected into a subspace. This processing not only preserves the channel multipath information as much as possible, but also reduces the data dimension. Based on the reduced dimension projected data, the ELM network is exploited to implement the fingerprint localization. Considering the limited performance of a single ELM network, multiple ELM networks are jointly optimized to improve the localization performance. The experimental results demonstrate that the proposed MuSpel-Fi algorithm has higher positioning accuracy than traditional ones.

End-to-end learning-based image dehazing methods tend to overdehaze or underdehaze in real scenes due to inefficient feature extraction and feature fusion. In this letter, we propose a multiscale supervision-guided context aggregation network (MSGCAN) based on two principles: improving feature extraction and enhancing feature mapping. To improve feature extraction, an attention-guided context aggregation (AGCA) module is adopted to merge context features extracted by several residual dense blocks (RDB). Moreover, we output these aggregated context features on each scale and form multiscale supervision to enhance feature mapping and ensure that the extracted features on each scale contain more realistic details. The experimental results show that the proposed MSGCAN performs better than other state-of-the-art dehazing methods in both synthetic and real-world scenes.

Abstract Single‐image dehazing is an important problem because it is a key prerequisite for most high‐level computer vision tasks. Traditional prior‐based methods adopt priors generated from clear images to restrain the atmospheric scattering model and then recover haze‐free images. However, these prior‐based methods always encounter over‐enhancement, such as halos and colour distortion. To solve this problem, many works use a convolutional neural network to retrieve original images. However, without priors as guidance, these learning‐based methods dehaze effectively in synthetic datasets but perform poorly in real scenes. Hence, in this paper, we propose a prior‐guided multiscale network for single‐image dehazing named PGMNet. Specifically, prior‐based methods are adopted to acquire dehazed images of the training dataset in advance and then send these dehazed images to a parameter‐shared encoder to form multiscale features. During the decoding process, these multiscale features are adopted to guide the prior‐guided multiscale network to recover more image details. Moreover, considering that these prior‐based dehazed images usually contain some over‐enhanced regions, a spatial attention guided feature aggregation module and squeeze‐and‐excitation module are adopted to alleviate colour distortion. The proposed PGMNet takes the advantage of prior‐based methods in real haze removal and provides superior performance compared with the state‐of‐the‐art methods on both synthetic and real‐world datasets.

An adaptive shadow detection algorithm is proposed to eliminate interference on object detection from the shadow. The algorithm uses three components in YUV colour space to identify shadow pixels from the candidate foreground. An adaptive threshold estimator is designed to improve shadow detection accuracy and adaptive capacity in various lighting conditions. This estimator uses edge detection method to obtain global texture, as well statistical calculations to obtain the thresholds. Algorithm has the characteristic of low complexity and little restraint; hence it is suitable for real time‐moving shadow detection in various lighting conditions. Experiment results show that this algorithm can obtain a high detection accuracy and the time‐assume is greatly shortened compared with other algorithms with similar accuracy.