Search Results (21)

The closed-loop testing of autonomous driving systems critically depends on large-scale libraries of diverse and realistic 3D vehicle assets, yet current pipelines still rely on labor-intensive modeling or multi-view capture, making efficient construction a key bottleneck. To overcome this bottleneck and enable convenient, cost-effective 3D asset generation, we propose a semantic prior-guided framework for accurate and robust vehicle point cloud reconstruction from casually captured single-view photographs. Our framework is built on a diffusion backbone but is fundamentally driven by two forms of prior knowledge: First, geometric and appearance priors from camera-aware image features, masks, and distance-transform maps are projected onto the evolving point cloud, compensating for the severe information loss in single-view inputs. Second, we introduce distillation-style regulators—pretrained neural networks that encode vehicle type and model semantics; they act as teacher networks that impose high-level constraints on the generated point clouds, transferring rich semantic knowledge and effectively regularizing the learning process. With these priors, our model infers vehicle-specific semantics from limited observations and reconstructs high-quality 3D point cloud assets. On the 3DRealCar++ dataset, our method clearly surpasses state-of-the-art point cloud baselines in both F-score and Chamfer Distance. Full article

Brain–computer interfaces (BCIs) based on motor imagery (MI) aim to convert electroencephalographic (EEG) activity into reliable device commands across users and recording setups. However, low signal-to-noise ratio and strong inter-subject variability still limit true “plug-and-play” deployment without lengthy calibration. To address these challenges, we propose a multi-branch convolutional long short-term memory (CNN-LSTM) architecture that jointly performs multi-scale temporal feature extraction and within-trial sequence modeling. The model employs four parallel 1D convolutional branches with distinct kernel sizes, each followed by an LSTM module and late fusion, combined with group normalization and supervision over sequences of sub-windows within each trial. We evaluate the approach on the EEG Motor Movement/Imagery (EEGMMI) dataset from PhysioNet under strictly subject-independent conditions, and on the ISLab-MI Dataset, a 32-channel wearable-EEG collection designed to assess cross-setup robustness. On EEGMMI, the network achieves up to 82.63% accuracy for binary left/right MI and 74.10% for a four-class task using 4 s trials under 5-fold cross-validation, outperforming an EEGNet-style baseline by 1–10% depending on class count and window length. Under a leave-one-subject-out protocol, the model attains 74.9% mean accuracy for a three-class MI task. Zero-shot transfer to ISLab-MI yields 64.60% and 63.02% accuracy in three- and four-class settings, respectively, while brief subject-specific fine-tuning using only 20% of each session improves performance to 81.38% and 73.48%. These findings show that combining multi-scale convolutional feature extraction with explicit sequence modeling and robust normalization yields accurate, data-efficient, and portable MI decoders suitable for practical BCI applications. Full article

Remote sensing provides vast unlabeled imagery at low cost, yet annotation remains expensive, making self-supervised learning (SSL) well suited to this domain. However, existing DINO-style SSL frameworks are not well suited to remote sensing imagery, where object extents vary substantially and standard multi-crop view generation often introduces cross-scale inconsistency. This issue is particularly severe for small objects and elongated structures, whose discriminative features can be lost under scale transformations. To address this limitation, we propose DINO-MS (DINO with multi-scale consistency), a scale-consistent SSL framework for remote sensing imagery. The key idea is to construct feature-aligned cross-scale local views and explicitly enforce prediction-level agreement among them. Specifically, DINO-MS first adopts a co-located multi-scale cropping strategy to sample local views from the same spatial location at different crop scales, and then introduces a local consistency loss that works jointly with the original DINO local-to-global objective. Extensive experiments on land-use classification and change detection benchmarks show that DINO-MS generally improves downstream transfer performance. Notably, on EuroSAT, it improves per-class accuracy from 80.60% to 87.80% for Highway and from 88.00% to 91.60% for River with DINO-MC, confirming its advantage for categories dominated by small objects. Full article

Exemplar-based image translation generates an output image by transferring appearance from a reference exemplar to a content image. Existing works only consider the local correspondences between two modalities, and ignore the global distributions in each modality, struggling to obtain fine-grained details with efficient computation. In this paper, we propose OTFormer, a multi-scale Optimal Transport transformer for exemplarbased image translation. We formulate cross-modal alignment as a multi-scale optimal transport problem, which progressively provides a globally coherent matching. In addition, we design a lightweight multi-scale fusion block to extract and fuse features efficiently. Experiments on CelebA-HQ and DeepFashion demonstrate that OTFormer improves both image fidelity and style adherence, while reducing model parameters by 62% and achieving faster inference compared with strong baselines. These results highlight OTguided global alignment as an effective and deployable solution for high-fidelity exemplarbased image translation. Full article

This manuscript presents a transferable and reproducible methodology for quantitative 3D tree-structure mensuration and transparent, rule-based decision support for tree risk management. The workflow integrates (i) Structure-from-Motion/Multi-View Stereo (SfM–MVS) reconstruction from multi-view imagery, (ii) independent referencing to ensure metric scaling and a consistent local frame, and (iii) point cloud analytics to derive branch-level geometric descriptors (e.g., base diameter, length, inclination, slenderness, and projected reach). A clear rule-based layer operationalizes Tree Risk Assessment Qualification (TRAQ)-style risk components and As Low As Reasonably Practicable (ALARP) principles to map geometry and exposure into auditable management recommendations (e.g., monitoring intervals, pruning/weight reduction, supplemental support, and exclusion-zone planning). To provide a real-data example, the demonstration uses the public Fuji-SfM apple orchard dataset, including three neighboring trees with partially overlapping crowns for tree instance extraction and subsequent TRAQ/ALARP scenarios on an outer tree. The proposed decision layer is intentionally based on external geometry and exposure; internal decay indicators and species-specific mechanical properties (e.g., Modulus of Elasticity (MOE), Modulus of Rupture (MOR)) are outside this demonstration and should be incorporated via complementary diagnostics in operational deployments. Full article

Domain-adaptive object detection under set-prediction paradigms remains challenging, as Hungarian matching is sensitive to domain shift and fixed pseudo-label thresholds cannot simultaneously handle class imbalance and scene variability. This paper presents SSRT-DETR, a semi-supervised, domain-adaptive framework built on the real-time detector RT-DETR. We adopt a mean teacher–student architecture with style-transferred images to jointly model source and target domains. To stabilize the assignment process during the early stages of cross-domain training, Domain-Aware Matching (DAM) is formulated to augment the Hungarian matching cost with a teacher-guided decoder-query consistency term. Leveraging the more stable EMA teacher representations, DAM guides early matching toward domain-consistent assignments and is gradually annealed to recover standard matching as training converges. In parallel, we introduce Class-/Scene-Adaptive Pseudo-Labeling (CAP) to address a key limitation of existing DAOD methods that rely on fixed or globally tuned pseudo-label thresholds, which struggle with class imbalance and scene-dependent difficulty under domain shift. CAP leverages per-class confidence statistics and multi-view consistency to adapt classification and IoU thresholds across classes and scenes, while temperature scaling and quality-weighted losses provide soft control over pseudo-label reliability. Experiments on standard benchmarks demonstrate the robustness of SSRT-DETR. On Cityscapes→Foggy Cityscapes, SSRT-DETR improves mAP@0.5 from 51.0 to 54.3. On KITTI→Cityscapes and Sim10K→Cityscapes, it achieves 67.3 AP and 64.9 AP on the car category, respectively, clearly outperforming the RT-DETR baseline while maintaining real-time efficiency. Notably, consistent gains are observed in rare categories and adverse weather scenarios, validating the effectiveness of the proposed DAM and CAP modules. Full article

AI-driven imaging is becoming central to crop monitoring, with proximal and unmanned aerial vehicle (UAV) platforms now routinely used for disease and stress detection, yield estimation, canopy structure, and fruit counting. Yet, as these models move from plots to farms, the main bottleneck is no longer raw accuracy but robustness under distribution shift. Systems trained in one field, season, cultivar, or sensor often fail when the scene, sensor, protocol, or timing changes in realistic ways. This review synthesizes recent advances on robustness and transferability in proximal and UAV imaging, drawing on a corpus of 42 core studies across field crops, orchards, greenhouse environments, and multi-platform phenotyping. Shift types are organized into four axes, namely scene, sensor, protocol, and time. The article also maps the empirical evidence on when RGB imaging alone is sufficient and when multispectral, hyperspectral, or thermal modalities can potentially improve robustness. This serves as a basis to synthesize acquisition and evaluation practices that often matter more than architectural tweaks, which include phenology-aware flight planning, radiometric standardization, metadata logging, and leave-one-field/season-out splits. Adaptation options are consolidated into a practical symptom/remedy roadmap, ranging from lightweight normalization and small target-set fine-tuning to feature alignment, unsupervised domain adaptation, style translation, and test-time updates. Finally, a benchmark and dataset agenda are outlined with emphasis on object-oriented splits, cross-sensor and cross-scale collections, and longitudinal datasets where the same fields are followed across seasons under different management regimes. The goal is to outline practices and evaluation protocols that support progress toward deployable and auditable systems, noting that such claims require standardized out-of-distribution testing and transparent reporting as emphasized in the benchmark specification and experiment suite proposed here. Full article

Cyberbullying and hate speech increasingly appear in multi-modal social media posts, where images and text are combined in diverse and fast changing ways across platforms. These posts differ in style, vocabulary and layout, and labeled data are sparse and noisy, which makes it difficult to train detectors that are both reliable and deployable under tight computational budgets. Many high performing systems rely on large vision language backbones, full parameter fine tuning, online retrieval or model ensembles, which raises training and inference costs. We present a parameter efficient cross-platform multi-modal transfer learning framework for cyberbullying and hateful content detection. Our framework has three components. First, we perform domain adaptive pretraining of a compact ViLT backbone on in domain image-text corpora. Second, we apply parameter efficient fine tuning that updates only bias terms, a small subset of LayerNorm parameters and the classification head, leaving the inference computation graph unchanged. Third, we use noise aware knowledge distillation from a stronger teacher built from pretrained text and CLIP based image-text encoders, where only high confidence, temperature scaled predictions are used as soft labels during training, and teacher models and any retrieval components are used only offline. We evaluate primarily on Hateful Memes and use IMDB as an auxiliary text only benchmark to show that the deployment aware PEFT + offline-KD recipe can still be applied when other modalities are unavailable. On Hateful Memes, our student updates only 0.11% of parameters and retain about 96% of the AUROC of full fine-tuning. Full article

Root-cause analysis (RCA) in large-scale microservice-based payment systems is challenging due to complex failure propagation along service dependencies, limited availability of labeled incident data, and heterogeneous service topologies across deployments. We propose ServiceGraph-FM, a pretrained graph-based model for RCA, where “foundation” denotes a self-supervised graph encoder pretrained on large-scale production cluster traces and then adapted to downstream diagnosis. ServiceGraph-FM introduces three components: (1) masked graph autoencoding pretraining to learn transferable service-dependency embeddings for cross-topology generalization; (2) a temporal relational diffusion module that models anomaly propagation as graph diffusion on dynamic service graphs (i.e., Laplacian-governed information flow with learnable edge propagation strengths); and (3) a causal attention mechanism that leverages multi-hop path signals to better separate likely causes from correlated downstream effects. Experiments on the Alibaba Cluster Trace and synthetic PayPal-style topologies show that ServiceGraph-FM outperforms state-of-the-art baselines, improving Top-1 accuracy by 23.7% and Top-3 accuracy by 18.4% on average, and reducing mean time to detection by 31.2%. In zero-shot deployment on unseen architectures, the pretrained model retains 78.3% of its fully fine-tuned performance, indicating strong transferability for practical incident management. Full article

Semantic segmentation of UAV–acquired RGB orthomosaics is a key component for quantifying vegetation cover and monitoring phenology in precision agriculture. This study evaluates a representative set of CNN–based architectures (U–Net, U–Net Xception–Style, SegNet, DeepLabV3+) and Transformer–based models (Swin–UNet/Swin–Transformer, SegFormer, and Mask2Former) under a unified and reproducible protocol. We propose a transfer–and–consolidation workflow whose performance is assessed not only through region–overlap and pixel–wise discrepancy metrics, but also via boundary–sensitive criteria that are explicitly linked to orthomosaic–scale vegetation–cover estimation by pixel counting under GSD (Ground Sample Distance) control. The experimental design considers a transfer scenario between morphologically related crops: initial training on Opuntia spp. (prickly pear), direct (“zero–shot”) inference on Agave salmiana, fine–tuning using only 6.84% of the agave tessellated set as limited target–domain supervision, and a subsequent consolidation stage to obtain a multi–species model. The evaluation integrates IoU, Dice, RMSE, pixel accuracy, and computational cost (time per image), and additionally reports the BF score and HD95 to characterize contour fidelity, which is critical when area is derived from orthomosaic–scale masks. Results show that Transformer-based approaches tend to provide higher stability and improved boundary delineation on Opuntia spp., whereas transfer to Agave salmiana exhibits selective degradation that is mitigated through low–annotation–cost fine-tuning. On Opuntia spp., Mask2Former achieves the best test performance (IoU 0.897 +/− 0.094; RMSE 0.146 +/− 0.002) and, after consolidation, sustains the highest overlap on both crops (IoU 0.894 +/− 0.004 on Opuntia and IoU 0.760 +/− 0.046 on Agave), while preserving high contour fidelity (BF score 0.962 +/− 0.102/0.877 +/− 0.153; HD95 2.189 +/− 3.447 px/8.458 +/− 16.667 px for Opuntia/Agave), supporting its use for final vegetation–cover quantification. Overall, the study provides practical guidelines for architecture selection under hardware constraints, a reproducible transfer protocol, and an orthomosaic–oriented implementation that facilitates integration into agronomic and remote–sensing workflows. Full article

Benefiting from the advantages of unmanned aerial vehicle (UAV) platforms such as low cost, rapid deployment capability, and miniaturization, the application of UAV-borne synthetic aperture radar (SAR) has developed rapidly. Utilizing a self-developed monostatic Miniaturized SAR (MiniSAR) system and a bistatic MiniSAR system, our team conducted multiple imaging missions over the same vehicle equipment display area at different times. However, system disparities and time-varying factors lead to a mismatch between the distributions of the training and test data. Additionally, small ground vehicle targets under complex background clutter exhibit limited size and weak scattering characteristics. These two issues pose significant challenges to the precise detection of small ground vehicle targets. To address these issues, this article proposes a domain-unified adaptive target detection framework (DUA-TDF). The approach consists of two stages: image-to-image translation and feature extraction and target detection. In the first stage, a multi-scale detail-aware CycleGAN (MSDA-CycleGAN) is proposed to align the source and target domains at the image level by achieving unpaired image style transfer while emphasizing both global structure and local details of the generated images. In the second stage, a cross-window axial self-attention target detection network (CWASA-Net) is proposed. This network employs a hybrid backbone centered on the cross-window axial self-attention mechanism to enhance feature representation, coupled with a convolution-based stacked cross-scale feature fusion network to strengthen multi-scale feature interaction. To validate the effectiveness and generalization capability of the proposed algorithm, comprehensive experiments are conducted on both self-developed monostatic/bistatic SAR datasets and public dataset. Experimental results demonstrate that our method achieves an mAP50 exceeding 90% in within-domain tests and maintains over 80% in cross-domain scenarios, demonstrating exceptional and robust detection performance as well as cross-domain adaptability. Full article

Recently, there has been significant progress in the area of semantic classification of water bodies at global scales with deep learning. For the key purposes of water inventory and change detection, advanced deep learning classifiers such as UNets and Vision Transformers have been shown to be both accurate and flexible when applied to large-scale, or even global, satellite image datasets from optical (e.g., Sentinel-2) and radar sensors (e.g., Sentinel-1). Most of this work is conducted with optical sensors, which usually have better image quality, but their obvious limitation is cloud cover, which is why radar imagery is an important complementary dataset. However, radar imagery is generally more sensitive to soil moisture than optical data. Furthermore, topography and wind-ripple effects can alter the reflected intensity of radar waves, which can induce errors in water classification models that fundamentally rely on the fact that water is darker than the surrounding landscape. In this paper, we develop a solution to the use of Sentinel-1 radar images for the semantic classification of water bodies that uses style transfer with multi-modal and multi-temporal image fusion. Instead of developing new semantic classification models that work directly on Sentinel-1 images, we develop a global style transfer model that produces synthetic Sentinel-2 images from Sentinel-1 input. The resulting synthetic Sentinel-2 imagery can then be classified with existing models. This has the advantage of obviating the need for large volumes of manually labeled Sentinel-1 water masks. Next, we show that fusing an 8-year cloud-free composite of the near-infrared band 8 of Sentinel-2 to the input Sentinel-1 image improves the classification performance. Style transfer models were trained and validated with global scale data covering the years 2017 to 2024, and include every month of the year. When tested against a global independent benchmark, S1S2-Water, the semantic classifications produced from our synthetic imagery show a marked improvement with the use of image fusion. When we use only Sentinel-1 data, we find an overall IoU (Intersection over Union) score of 0.70, but when we add image fusion, the overall IoU score rises to 0.93. Full article

To tackle phenotypic variability and detection accuracy issues of tea shoots in open-air gardens due to lighting and varietal differences, this study proposes Tea CycleGAN and a data augmentation method. It combines multi-scale image style transfer with spatial consistency dataset generation. Using Longjing 43 and Zhongcha 108 as cross-domain objects, the generator integrates SKConv and a dynamic multi-branch residual structure for multi-scale feature fusion, optimized by an attention mechanism. A deep discriminator with more conv layers and batch norm enhances detail discrimination. A global–local framework trains on 600 × 600 background and 64 × 64 tea shoots regions, with a restoration-paste strategy to preserve spatial consistency. Experiments show Tea CycleGAN achieves FID scores of 42.26 and 26.75, outperforming CycleGAN. Detection using YOLOv7 sees mAP rise from 73.94% to 83.54%, surpassing Mosaic and Mixup. The method effectively mitigates lighting/scale impacts, offering a reliable data augmentation solution for tea picking. Full article

Artistic style transfer aims to transfer the style of an artwork to a photograph while maintaining its original overall content. Although current style-transfer methods have achieved promising results when processing photorealistic images, they often struggle with brushstroke preservation in artworks, especially in styles such as oil painting and pointillism. In such cases, the extracted style and content features tend to include redundant information, leading to issues such as blurred edges and a loss of fine details in the transferred images. To address this problem, this paper proposes a multi-scale general style-transfer network based on diffusion models. The proposed network consists of a coarse style-transfer module and a refined style-transfer module. First, the coarse style-transfer module is designed to perform mainstream style-transfer tasks more efficiently by operating on downsampled images, enabling faster processing with satisfactory results. Next, to further enhance edge fidelity, a refined style-transfer module is introduced. This module utilizes a segmentation component to generate a mask of the main subject in the image and performs edge-aware refinement. This enhances the fusion between the subject’s edges and the target style while preserving more detailed features. To improve overall image quality and better integrate the style along the content boundaries, the output from the coarse module is upsampled by a factor of two and combined with the subject mask. With the assistance of ControlNet and Stable Diffusion, the model performs content-aware edge redrawing to enhance the overall visual quality of the stylized image. Compared with state-of-the-art style-transfer methods, the proposed model preserves more edge details and achieves more natural fusion between style and content. Full article

Cross-modal image registration for unmanned aerial vehicle (UAV) platforms presents significant challenges due to large-scale deformations, distinct imaging mechanisms, and pronounced modality discrepancies. This paper proposes a novel multi-scale cascaded registration network based on style transfer that achieves superior performance: up to 67% reduction in mean squared error (from 0.0106 to 0.0068), 9.27% enhancement in normalized cross-correlation, 26% improvement in local normalized cross-correlation, and 8% increase in mutual information compared to state-of-the-art methods. The architecture integrates a cross-modal style transfer network (CSTNet) that transforms visible images into pseudo-infrared representations to unify modality characteristics, and a multi-scale cascaded registration network (MCRNet) that performs progressive spatial alignment across multiple resolution scales using diffeomorphic deformation modeling to ensure smooth and invertible transformations. A self-supervised learning paradigm based on image reconstruction eliminates reliance on manually annotated data while maintaining registration accuracy through synthetic deformation generation. Extensive experiments on the LLVIP dataset demonstrate the method’s robustness under challenging conditions involving large-scale transformations, with ablation studies confirming that style transfer contributes 28% MSE improvement and diffeomorphic registration prevents 10.6% performance degradation. The proposed approach provides a robust solution for cross-modal image registration in dynamic UAV environments, offering significant implications for downstream applications such as target detection, tracking, and surveillance. Full article