CVT (Centroidal Voronoi Tessellation)-based remeshing optimizes mesh quality via the Voronoi-Delaunay framework, optimizing vertex distribution and generating regular triangles. Current CVT-based approaches can fall into two categories. The former are exact methods, such as Geodesic CVT and Restricted Voronoi Diagrams(RVD), which ensure high quality but require significant computation. The latter are approximate methods, which reduce computational complexity yet compromise quality. To address this tradeoff, we propose a CVT-based surface remeshing method that balances optimization between quality and efficiency via curvature-adaptive multi-clipping of 3D Centroidal Voronoi cells using original surface facets. The core idea of the method is that we adaptively adjust the number of clipping times according to local curvature, and use the angular relationship between the normal vectors of neighboring facets to represent the magnitude of local curvature. Experimental results demonstrate the effectiveness of our method.

Abstract Point cloud normal estimation underpins many 3D vision and graphics applications. Precise normal estimation in regions of sharp curvature and high‐frequency variation remains a major bottleneck; existing learning‐based methods still struggle to isolate fine geometry details under noise and uneven sampling. We present FAHNet, a novel frequency‐aware hierarchical network that precisely tackles those challenges. Our Frequency‐Aware Hierarchical Geometry (FAHG) feature extraction module selectively amplifies and merges cross‐scale cues, ensuring that both fine‐grained local features and sharp structures are faithfully represented. Crucially, a dedicated Frequency‐Aware geometry enhancement (FA) branch intensifies sensitivity to abrupt normal transitions and sharp features, preventing the common over‐smoothing limitation. Extensive experiments on synthetic benchmarks (PCPNet, FamousShape) and real‐world scans (SceneNN) demonstrate that FAHNet outperforms state‐of‐the‐art approaches in normal estimation accuracy. Ablation studies further quantify the contribution of each component, and downstream surface reconstruction results validate the practical impact of our design.

Autonomous, accurate, and dynamic 3-D reconstruction for wide-area environments is crucial for unmanned aerial vehicle monitoring and rescue tasks, however, when conducted in an unknown complex terrain, the reconstruction result obtained from a single flight suffers poor quality. In this article, we present an Active Iterative Optimization framework for trajectory planning and visual reconstruction. Firstly, the trajectory is planned under the photogrammetric constraints based on rough terrain. Due to the visual field deviation caused by pose error during actual flight, the view loss evaluation is established and keyframes are selected to conduct 3-D reconstruction. A comprehensive metric is designed to quantitatively evaluate reconstruction effect without ground truth. The point cloud is then rasterized and divided into normal or low-scoring region according to the evaluation metric. In the next iteration, trajectory is replanned in low-scoring region to purposefully optimize the point cloud of local area. Thus the reconstruction result can be iteratively optimized. We validated the effectiveness of the proposed framework in simulation and physical experiments.

Differentiable Architecture Search (DARTS) has attracted considerable attention as a gradient-based neural architecture search method. Since the introduction of DARTS, there has been little work done on adapting the action space based on state-of-art architecture design principles for CNNs. In this work, we aim to address this gap by incrementally augmenting the DARTS search space with micro-design changes inspired by ConvNeXt and studying the trade-off between accuracy, evaluation layer count, and computational cost. We introduce the Pseudo-Inverted Bottleneck Conv (PIBConv) block intending to reduce the computational footprint of the inverted bottleneck block proposed in ConvNeXt. Our proposed architecture is much less sensitive to evaluation layer count and outperforms a DARTS network with similar size significantly, at layer counts as small as 2. Furthermore, with less layers, not only does it achieve higher accuracy with lower computational footprint (measured in GMACs) and parameter count, GradCAM comparisons show that our network can better detect distinctive features of target objects compared to DARTS. Code is available from https://github.com/mahdihosseini/PIBConv.

Abstract Triangular meshes of superior quality are important for geometric processing in practical applications. Existing approximative CVT‐based remeshing methodology uses planar polygonal facets to fit the original surface, simplifying the computational complexity. However, they usually do not consider surface curvature. Topological errors and outliers can also occur in the close sheet surface remeshing, resulting in wrong meshes. With this regard, we present a novel method named PowerRTF, an extension of the restricted tangent face (RTF) in conjunction with the power diagram, to better approximate the original surface with curvature adaption. The idea is to introduce a weight property to each sample point and compute the power diagram on the tangent face to produce area‐controlled polygonal facets. Based on this, we impose the variable‐capacity constraint and centroid constraint to the PowerRTF, providing the trade‐off between mesh quality and computational efficiency. Moreover, we apply a normal verification‐based inverse side point culling method to address the topological errors and outliers in close sheet surface remeshing. Our method independently computes and optimizes the PowerRTF per sample point, which is efficiently implemented in parallel on the GPU. Experimental results demonstrate the effectiveness, flexibility, and efficiency of our method.

Antepartum cardiotocography (CTG) monitoring is a crucial screening tool widely utilized to evaluate fetal wellbeing. However, the complexity and non-linearity of CTG usually result in inter-observer and intra-observer variability in a visual CTG interpretation using clinical guidelines. In this paper, a fuzzy C-means clustering based adaptive neuro-fuzzy inference system (FCM-ANFIS) was proposed to automatically classify CTG for antenatal fetal monitoring. Data visualization and spearman correlation analysis were implemented to select CTG features. Then, the fuzzy space was partitioned by using fuzzy Cmeans clustering algorithm, and the adjustment parameters were adjusted through the self-learning mechanism of neural networks and least squares algorithm. The experimental results show that the fuzzy space partition based on FCM clustering could improve the performance of ANFIS, and the proposed FCM-ANFIS model outperforms the state-of-the-art automatic classification of CTG models. In conclusion, the proposed FCM-ANIFIS model has promising learning ability and adaptability for the complexity and uncertainty of antenatal CTG interpretation.

A few recent developments have demonstrated that deep-learning-based solutions can outperform traditional algorithms for automated pavement crack detection. In this paper, an efficient deep network called CrackNet-V is proposed for automated pixel-level crack detection on 3D asphalt pavement images. Compared with the original CrackNet, CrackNet-V has a deeper architecture but fewer parameters, resulting in improved accuracy and computation efficiency. Inspired by CrackNet, CrackNet-V uses invariant spatial size through all layers such that supervised learning can be conducted at pixel level. Following the VGG network, CrackNet-V uses 3 × 3 size of filters for the first six convolutional layers and stacks several 3 × 3 convolutional layers together for deep abstraction, resulting in reduced number of parameters and efficient feature extraction. CrackNet-V has 64113 parameters and consists of ten layers, including one pre-process layer, eight convolutional layers, and one output layer. A new activation function leaky rectified tanh is proposed in this paper for higher accuracy in detecting shallow cracks. The training of CrackNet-V was completed after 3000 iterations, which took only one day on a GeForce GTX 1080Ti device. According to the experimental results on 500 testing images, CrackNet-V achieves a high performance with a Precision of 84.31%, Recall of 90.12%, and an F-1 score of 87.12%. It is shown that CrackNet-V yields better overall performance particularly in detecting fine cracks compared with CrackNet. The efficiency of CrackNet-V further reveals the advantages of deep learning techniques for automated pixel-level pavement crack detection.

In dynamic-traffic Wavelength Division Multiplexing (WDM) networks, it is desirable to guarantee each optical circuit a minimum Optical Signal-to-Noise Ratio (OSNR) throughout the circuit's entire lifetime. The objective of this paper is to increase such guaranteed OSNR, which in turn enables bandwidth-efficient modulation formats to be applied to the circuit signals. The guaranteed OSNR is improved by applying OSNR-driven Spectrum Assignment (OSA) constraints to each fiber link in the network, based on its forecast traffic volume (offered load). Applying OSA constraints to fiber links has two conflicting outcomes. On the one hand, circuit requests may experience higher blocking probabilities due to the lack of usable spectrum. On the other hand, the adverse effects of medium nonlinearity may be contained, thus yielding improved circuit OSNR. Two OSA constraints are described and their favorable impact on circuit OSNR (up to 2.75 dB gain) is estimated by using a Gaussian Noise (GN) model and an event-driven simulation of two WDM networks supporting dynamic circuits.

Distributed and global power control strategies have been successfully applied to cascade of optical amplifiers with the aim of improving circuit OSNR values in WDM networks. For these or similar strategies to achieve their objectives, accurate estimators of the EDFA output power (EOP) are of the essence. These EOP estimators must take into account the amplifier characteristics including unequal power values of the input signals, frequency of each signal and desired target gain for each amplifier. The scope of this paper is to propose and assess the accuracy of three EOP efficient linear estimators. The EOP estimators are part of a framework comprising four modules, which are simple to implement and flexible to use. The proposed estimators are efficient in that they compute EOP estimates in seconds and require only a limited set of EDFA measured data.