Local Geometric Features Research Articles

Orientation-Preserving Spectral Correspondence for 3D Shape Analysis

Abstract In this article, the authors present an orientation-preserving spectral correspondence for three-dimensional (3D) shape analysis, which is robust and efficient for topological and deformable changes, even for non-isometric shapes. Our technique introduces an optimal spectral representation by combining the eigendecomposition with principal components analysis (PCA) to the heat kernel Laplacian matrix, and we further propose an efficient symmetry detection method based on so-called dominant eigenfunctions. Finally, a 3D descriptor encoding intrinsic symmetry structure and local geometric feature is constructed which effectively reveals the consistent structure between the deformable shapes. Consequently, sufficient orientation-preserving correspondence can be established in our embedding space. Experimental results showed that our method produces stable matching results in comparison with state-of-the-art methods.

Spontaneous object-location memory based on environmental geometry is impaired by both hippocampal and dorsolateral striatal lesions.

We examined the role of the hippocampus and the dorsolateral striatum in the representation of environmental geometry using a spontaneous object recognition procedure. Rats were placed in a kite-shaped arena and allowed to explore two distinctive objects in each of the right-angled corners. In a different room, rats were then placed into a rectangular arena with two identical copies of one of the two objects from the exploration phase, one in each of the two adjacent right-angled corners that were separated by a long wall. Time spent exploring these two objects was recorded as a measure of recognition memory. Since both objects were in different locations with respect to the room (different between exploration and test phases) and the global geometry (also different between exploration and test phases), differential exploration of the objects must be a result of initial habituation to the object relative to its local geometric context. The results indicated an impairment in processing the local geometric features of the environment for both hippocampus and dorsolateral striatum lesioned rats compared with sham-operated controls, though a control experiment showed these rats were unimpaired in a standard object recognition task. The dorsolateral striatum has previously been implicated in egocentric route-learning, but the results indicate an unexpected role for the dorsolateral striatum in processing the spatial layout of the environment. The results provide the first evidence that lesions to the hippocampus and dorsolateral striatum impair spontaneous encoding of local environmental geometric features.

Built-up area detection from high resolution remote sensing images using geometric features

Based on characteristic of High Spatial Resolution(HSR) image, an adaptive local geometric invariant feature extraction method based on Gabor transform and Tensor Voting(TV) is proposed and applied to building area extraction from HSR remote sensing images.First, image geometric features are analyzed. Second, the feasibility of extracting built-up area using local feature points and probability density estimation are determined. This study provides specific methods and steps. First, considering the abundant geometric features of high spatial resolution remote sensing images, multi-scale and multi-direction Gabor filter banks are adopted in detecting the singularity of images contain building area. In order to extract edge information of buildings, only real part Gabor coefficients are used. Meanwhile, we also measured the influence of Gabor kernels of different size on geometric feature extraction experiments, and the optimal scale parameter intervals for geometric feature extraction of remote sensing images with different spatial resolutions are thus determined. Second, it is a fact that the response of Gabor filter at each pixel is a measure for orientation certainty, thus, we introduce the orientation tensor which represents an ideal direction in the direction of the unit vector perpendicular to frequency. By weighting the orientation tensor with Gabor coefficients ans summing over it to complete the information fusion, the resulting tensor givens an estimate for local orientation and orientation uncertainty at image position. The tensors are then used as an initial estimate for global context refinement using tensor voting and the points are classified based on the their likelihoods of being part of a feature type, non-maximal suppression is used to extract geometric features. A key advantage of combining the Gabor filtering and tensor voting is that it eliminates the need for any thresholds therefore removing any data dependencies. In order to achieve a reliable extraction of local invariant feature such as corners from built-up area, three criterions are further proposed to refine the first stage result. Finally, each local building corner indicates a building be detected in image. However, only one of them is not sufficient alone to detecting a building. In fact, the more local points a buildings has, the more probable its detection becomes. Based on that fact, a probability density estimate method is generated to describe the probability that the pixel belongs to the building area, and the Otsu method is used to automatically extract the polygon area of the residential area. Experiments were carried out using image data sets such as Google and GF-2 with resolution higher than 1 meter, results showing that the proposed method can achieve higher accuracy in building area detection compared with the state-of-the-art corner detection algorithm such as Harris corner and High-speed corner detect method.

LPD-AE: Latent Space Representation of Large-Scale 3D Point Cloud

The effective latent space representation of point cloud provides a foremost and fundamental manner that can be used for challenging tasks, including point cloud based place recognition and reconstruction, especially in large-scale dynamic environments. In this paper, we present a novel deep neural network, LPD-AE(Large-scale Place Description AutoEncoder Network), to obtain meaningful local and contextual features for the generation of latent space from 3D point cloud directly. The encoder network constructs the discriminative global descriptors to realize high accuracy and robust place recognition, which contributed by extracting the local neighbor geometric features and aggregating neighborhood relationships both in feature space and physical space. The decoder network performs hierarchical reconstruction on coarse key points and ultimately produce dense point clouds, which shows that it is capable of reconstructing a full point cloud frame from a single compact but high dimensional descriptor. Our proposed network demonstrates performance that is comparable to the state-of-the-art approaches. With the benefit of the LPD-AE, many computationally complex tasks that rely directly on point clouds can be effortlessly conducted on latent space with lower memory costs, such as relocalization, loop closure detection, and map compression reconstruction. Comprehensive validations on Oxford RobotCar dataset, KITTI dataset, and our freshly collected dataset, which contains multiple trials of repeated routes in different weather and at different times, manifest its potency for real robotic and self-driving implementation. The source code is available at https://github.com/Suoivy/LPD-AE .

New method for identification of persons using geometry foot outline

In this study we try to improve existing recognition rates using plantar surface of foot to extract 19 local geometrical features based on anatomical points and axes, in order to extract informations and relationships required to perform a biometric system for identifying individuals. Before the primary data collection, a pilot study was conducted to assess the precision of stature acquisition and foot measurements. A computer database has been constructed taking measurements from feet pictures of 102 volunteers using the measuring tool in Photofilter. Using the measurements with ±0.38 like error margin and searching through it shows that each foot is unique. This study also showed that each measurement parameter's has an identification percentage and the highest percentages were noted for the parallel and diagonal lengths measurements. To validate our database and our results we tried to study the relation between the stature and some foot measurements using a linear equation of regression.

New method for identification of persons using geometry foot outline

In this study we try to improve existing recognition rates using plantar surface of foot to extract 19 local geometrical features based on anatomical points and axes, in order to extract informations and relationships required to perform a biometric system for identifying individuals. Before the primary data collection, a pilot study was conducted to assess the precision of stature acquisition and foot measurements. A computer database has been constructed taking measurements from feet pictures of 102 volunteers using the measuring tool in Photofilter. Using the measurements with ±0.38 like error margin and searching through it shows that each foot is unique. This study also showed that each measurement parameter's has an identification percentage and the highest percentages were noted for the parallel and diagonal lengths measurements. To validate our database and our results we tried to study the relation between the stature and some foot measurements using a linear equation of regression.

GGM-Net: Graph Geometric Moments Convolution Neural Network for Point Cloud Shape Classification

Geometric feature acts as an important role in point cloud shape classification tasks. Previous methods have proved that the geometric information of point clouds effectively improves the classification accuracy. Mo-Net firstly introduced geometric moments into point cloud shape classification, which, to fit the form of second order geometric moments, extends the number of input channels from three to nine. Unfortunately, similar to PointNet, Mo-Net cannot capture the local structures. To address this issue, we propose a graph geometric moments convolution neural network (GGM-Net), which learns local geometric features from geometric moments representation of a local point set. The core module of the GGM-Net is to learn features from geometric moments (termed as GGM convolution). Specifically, the GGM convolution learns point features and local features from the first and second order geometric moments of points and its local neighbors, respectively, and then combines these features by using an addition operation. In this way, a geometrical local representation about points is obtained, which leads to much surface geometry awareness and robustness. Equipped with the GGM convolution, GGM-Net, a simple end-to-end architecture, is developed to achieve a competitive accuracy on the benchmark dataset ModelNet40 and perform more efficiently in terms of memory and computational complexity.

Predicting the impacts of mutations on protein-ligand binding affinity based on molecular dynamics simulations and machine learning methods

PurposeMutation-induced variation of protein-ligand binding affinity is the key to many genetic diseases and the emergence of drug resistance, and therefore predicting such mutation impacts is of great importance. In this work, we aim to predict the mutation impacts on protein-ligand binding affinity using efficient structure-based, computational methods. MethodsRelying on consolidated databases of experimentally determined data we characterize the affinity change upon mutation based on a number of local geometrical features and monitor such feature differences upon mutation during molecular dynamics (MD) simulations. The differences are quantified according to average difference, trajectory-wise distance or time-vary differences. Machine-learning methods are employed to predict the mutation impacts using the resulting conventional or time-series features. Predictions based on estimation of energy and based on investigation of molecular descriptors were conducted as benchmarks. ResultsOur method (machine-learning techniques using time-series features) outperformed the benchmark methods, especially in terms of the balanced F1 score. Particularly, deep-learning models led to the best prediction performance with distinct improvements in balanced F1 score and a sustained accuracy. ConclusionOur work highlights the effectiveness of the characterization of affinity change upon mutations. Furthermore, deep-learning techniques are well designed for handling the extracted time-series features. This study can lead to a deeper understanding of mutation-induced diseases and resistance, and further guide the development of innovative drug design.

A Convolutional Neural Network for Point Cloud Instance Segmentation in Cluttered Scene Trained by Synthetic Data Without Color

3D Instance segmentation is a fundamental task in computer vision. Effective segmentation plays an important role in robotic tasks, augmented reality, autonomous driving, etc. With the ascendancy of convolutional neural networks in 2D image processing, the use of deep learning methods to segment 3D point clouds receives much attention. A great convergence of training loss often requires a large amount of human-annotated data, while making such a 3D dataset is time-consuming. This paper proposes a method for training convolutional neural networks to predict instance segmentation results using synthetic data. The proposed method is based on the SGPN framework. We replaced the original feature extractor with “dynamic graph convolutional neural networks” that learned how to extract local geometric features and proposed a simple and effective loss function, making the network more focused on hard examples. We experimentally proved that the proposed method significantly outperforms the state-of-the-art method in both Stanford 3D Indoor Semantics Dataset and our datasets.

A Cartoon+Texture Image Decomposition Variational Model Based on Preserving the Local Geometric Characteristics

This paper is devoted to decomposing one given image into the cartoon and texture. The cartoon parts correspond to the main large objects in the image, and the textural parts contain fine scale details. Mathematically, these two components should belong to the different function spaces corresponding to different norms. In most variational models of image decomposition, the cartoon parts are described by the total variation norm and the texture parts are characterized by its dual norm. Using these methods, the cartoon and texture can be separated from the original image. Some structures, such as edges, are well preserved in the cartoon image. However, as far as we know, none of these methods considers the geometric characteristics information of the texture position in the separated cartoon images. That is, the geometrical information in those separated cartoon images may have been destroyed, which can be seen from the experiments in this paper. To maintain these geometrical characteristics, we introduce one smoothed vector field and let it approximate to the negative gradient direction of the cartoon image, and then for the smoothed vector field, we let it belong to the Lebesgue space, thus a new variational decomposition model is proposed. The corresponding alternating direction method is discussed in detail. Experimental results are reported to show the visual qualities compared with other methods.

Point-PlaneNet: Plane kernel based convolutional neural network for point clouds analysis

Point cloud is accepted as an adequate representation for 3D data and most 3D sensors have the ability to generate this data. Due to point cloud's irregular format, analyzing this data using deep learning algorithms is quite challenging. In this paper, a new convolutional neural network, called Point-PlaneNet, is proposed that uses the concept of the distance between points and planes in order to exploit spatial local correlations. In the proposed method an alternative simple local operation, called PlaneConv, is introduced which can extract local geometric features from point clouds by learning a set of planes in Rn space. Our network takes raw point clouds as input and therefore avoids the need to transform point clouds to images or volumes. PlaneConv has a simple theoretical analysis and is easy to incorporate into deep learning models to improve their performance. In order to evaluate the proposed method for classification, part segmentation and scene semantic segmentation tasks, it has been applied on four Datasets: ModelNet-40, MNIST, ShapeNet-Part and S3DIS. The experimental results show the acceptable performance of the proposed method compared to previous approaches in all tasks.

Evaluating Local Geometric Feature Representations for 3D Rigid Data Matching.

Local geometric descriptors act as an essential component for 3D rigid data matching. A rotational invariant local geometric descriptor usually consists of two components: local reference frame (LRF) and feature representation. However, existing evaluation efforts have mainly been paid on the LRF or the overall descriptor and the quantitative comparison of feature representations remains unexplored. This paper fills the gap by comprehensively evaluating nine state-of-the-art local geometric feature representations. In particular, our evaluation assesses feature representations based on ground-truth LRFs such that the ranking of tested methods is more convincing as compared with existing studies. The experiments are deployed on six standard datasets with various application scenarios (shape retrieval, point cloud registration, and object recognition) and data modalities (LiDAR, Kinect, and Space Time) as well as perturbations including Gaussian noise, shot noise, data decimation, clutter, occlusion, and limited overlap. The evaluated terms cover the major concerns for a feature representation, e.g., distinctiveness, robustness, compactness, and efficiency. The outcomes present interesting findings that may shed new light on this community and provide complementary perspectives to existing evaluations on the topic of local geometric feature description. A summary of evaluated methods regarding their peculiarities is finally presented to guide real-world applications and new descriptor crafting.

Texture target classification with CLBP and local geometric features

For the problems of needing pre-training and poor robustness to rotation and illumination changes of various improved algorithms based on local binary pattern (LBP), this paper presents a new texture classification algorithm by integrating the completed local binary pattern (CLBP) and the local geometric invariant features of the image surface. In our algorithm, the local geometric invariant features are first computed. Then the computed results are further quantified and encoded to make combination with the CLBP histogram. The proposed algorithm can extract image macroscopic and microscopic features simultaneously, and it has the properties of not significantly increasing feature dimension, without pre-training, and invariance to the rotation and illumination changes. Experimental verifications are conducted on two standard texture databases, and the results demonstrate that the proposed algorithm outperforms the comparative classification algorithms in classification accuracy and robustness.

Unsupervised Dimensionality Reduction for Hyperspectral Imagery via Local Geometric Structure Feature Learning

Hyperspectral images (HSIs) possess a large number of spectral bands, which easily lead to the curse of dimensionality. To improve the classification performance, a huge challenge is how to reduce the number of spectral bands and preserve the valuable intrinsic information in the HSI. In this letter, we propose a novel unsupervised dimensionality reduction method called local neighborhood structure preserving embedding (LNSPE) for HSI classification. At first, LNSPE reconstructs each sample with its spectral neighbors and obtains the optimal weights for constructing the adjacency graph by modifying its loss function. Then, to discover the scatter information of the training samples, LNSPE minimizes the scatter between the pixels and the corresponding neighbors and maximizes the total scatter of the HSI data. Finally, it incorporates the scatter information and the dual graph structure to enhance the aggregation of the HSI. As a result, LNSPE can effectively reveal the intrinsic structure and improve the classification performance of the HSI data. The experimental results on two real hyperspectral data sets exhibit the efficiency and superiority of LNSPE to some state-of-the-art methods.

Numerical Modeling of Turbine Blade Cooling for Aero Engine Applications with the Use of Surrogate Models

The optimization of the thermodynamic cycles of aero engines has always been in the main targets of engineering efforts for environmental and economic reasons. The aero engine thermodynamic cycle is significantly affected by the selection of the cycle maximum temperature, which at the same time, should be high enough to achieve increased cycle efficiency but also be always kept within turbine blades material temperature limits to ensure turbine blades endurance and integrity. For these reasons, cooling techniques are used where a part of the compressor discharge air is usually used as a cooling medium for the turbine blades. Since this part of air does not participate in the heat addition process inside the combustion chamber its amount should be carefully estimated in order to simultaneously provide cooling air protection for the turbine blades and avoid loss of potential turbine work. Towards this direction, numerical tools such as the one presented in this work are developed. The present work is focused on the development of a surrogate model network for the calculation of turbine blade cooling for aero engine applications. In the analysis the turbine blade is modelled as a heat exchange unit composed of a number of interconnected sub-units. In each sub-unit the effect of turbine blade local geometrical features, such as inner channels length and hydraulic diameter, on heat transfer and pressure losses was incorporated through the use of literature based correlations. The investigations were focused on typical turbine blade conditions for recuperative aero engine applications as presented in Salpingidou et al. (2017). For the cooling mass flow calculations the suggestions of Young and Wilcock (2002a and 2002b) and Wilcock et al. (2005) were taken into consideration. Furthermore, after the development of the surrogate model network, the effect of incorporating heat transfer enhancers on inner flow turbine blade geometry was numerically assessed for two heat transfer augmentation surfaces, such as the ones presented in the work of Alam and Kim (2018), using experimental data available in literature describing their heat transfer and pressure loss characteristics targeting the identification of configurations with promising effect on the aero engine.

On shortened 3D local binary descriptors

The wide-spread mobile systems nowadays desire ultra lightweight local geometric features to accomplish tasks relying on correspondences. Nonetheless, most existing 3D local feature descriptors, though shown to be distinctive and robust, still are real-valued and/or high-dimensional. Accordingly, this paper conducts a comparative study on current bit-selection methods with a focus on shortening 3D local binary descriptors. By analyzing several bit-selection techniques, we develop and evaluate various approaches to obtain a shortened version of a state-of-the-art feature remaining discriminative and robust. Through extensive experiments on four standard datasets with different data modalities (e.g., LiDAR and Kinect) and application scenarios (e.g., 3D object retrieval, 3D object recognition, and point cloud registration), we show that a small subset of representative bits are sufficient to achieve promising feature matching results as the initial descriptor. Moreover, the shortened binary descriptors still hold competitive or better distinctiveness and robustness compared to several state-of-the-art real-valued descriptors, e.g., spin image, SHOT, and RoPS, albeit being dramatically more efficient to match and store. Key to the foreseen research trend of local geometric feature description is dealing with compact binary descriptors; thus, our work may pave the way for this new research direction.

Combining active learning and local patch alignment for data-driven facial animation with fine-grained local detail

Data-driven facial animation has attracted considerable attention from both academic and industrial communities in recent years. Typically, the motion data used to animate the faces are derived from either 3D or 2D facial features whose positions on the face are determined according to experience. There still lacks an automatic approach to determine the optimal positions of the features to face deformation, and current face deformation methods are incapable of providing fine-grained local geometric characteristics. This paper proposes a novel scheme for face animation in which an active learning method based on Locally Linear Reconstruction algorithm is exploited to determine the optimal feature positions on the face for face deformation, and the Semi-Supervised Local Patch Alignment algorithm is subsequently used to deform the face with the selected features according to the optimal feature positions. The active learning model can be solved by a sequential approach, and the Semi-Supervised Local Patch Alignment model can be addressed by a least-square method. Experimental results on various types of faces demonstrate the superiority of the proposed scheme to existing approaches in both feature points selection and fine-grained local characteristics preservation.

Accuracy Assessment of Deep Learning Based Classification of LiDAR and UAV Points Clouds for DTM Creation and Flood Risk Mapping

Digital elevation model (DEM) has been frequently used for the reduction and management of flood risk. Various classification methods have been developed to extract DEM from point clouds. However, the accuracy and computational efficiency need to be improved. The objectives of this study were as follows: (1) to determine the suitability of a new method to produce DEM from unmanned aerial vehicle (UAV) and light detection and ranging (LiDAR) data, using a raw point cloud classification and ground point filtering based on deep learning and neural networks (NN); (2) to test the convenience of rebalancing datasets for point cloud classification; (3) to evaluate the effect of the land cover class on the algorithm performance and the elevation accuracy; and (4) to assess the usability of the LiDAR and UAV structure from motion (SfM) DEM in flood risk mapping. In this paper, a new method of raw point cloud classification and ground point filtering based on deep learning using NN is proposed and tested on LiDAR and UAV data. The NN was trained on approximately 6 million points from which local and global geometric features and intensity data were extracted. Pixel-by-pixel accuracy assessment and visual inspection confirmed that filtering point clouds based on deep learning using NN is an appropriate technique for ground classification and producing DEM, as for the test and validation areas, both ground and non-ground classes achieved high recall (>0.70) and high precision values (>0.85), which showed that the two classes were well handled by the model. The type of method used for balancing the original dataset did not have a significant influence in the algorithm accuracy, and it was suggested not to use any of them unless the distribution of the generated and real data set will remain the same. Furthermore, the comparisons between true data and LiDAR and a UAV structure from motion (UAV SfM) point clouds were analyzed, as well as the derived DEM. The root mean square error (RMSE) and the mean average error (MAE) of the DEM were 0.25 m and 0.05 m, respectively, for LiDAR data, and 0.59 m and –0.28 m, respectively, for UAV data. For all land cover classes, the UAV DEM overestimated the elevation, whereas the LIDAR DEM underestimated it. The accuracy was not significantly different in the LiDAR DEM for the different vegetation classes, while for the UAV DEM, the RMSE increased with the height of the vegetation class. The comparison of the inundation areas derived from true LiDAR and UAV data for different water levels showed that in all cases, the largest differences were obtained for the lowest water level tested, while they performed best for very high water levels. Overall, the approach presented in this work produced DEM from LiDAR and UAV data with the required accuracy for flood mapping according to European Flood Directive standards. Although LiDAR is the recommended technology for point cloud acquisition, a suitable alternative is also UAV SfM in hilly areas.

Data-Driven Research on the Matching Degree of Eyes, Eyebrows and Face Shapes.

There is a close relationship between the attractiveness of the face and the facial features. The shape of the facial features determines the level of attractiveness, in which the eyes and eyebrows are particularly vital. In this article, we proposed a method to study the facial attractiveness by combining global face shape and local geometric features of eye and eyebrow and using computer big data analysis for assistance. Firstly, we collected 300 images of East Asian female and use machine learning methods to evaluate the attractiveness scores of face images. Secondly, geometric models were constructed separately for the eyebrows and the eyes to obtain their geometric and shape features. Correlation analysis was performed on the obtained data to study their shape matching of different facial attractiveness rating levels. Finally, the relationship between the shape of face and eyebrows was analyzed by combining the facial ratio and the geometric features of the eyebrows. The research in this article can provide reference for medical and beauty institutions and women’s makeup, and further study in the field of facial aesthetic analysis based on geometric features.

Anomaly Detection Based on Mining Six Local Data Features and BP Neural Network

Key performance indicators (KPIs) are time series with the format of (timestamp, value). The accuracy of KPIs anomaly detection is far beyond our initial expectations sometimes. The reasons include the unbalanced distribution between the normal data and the anomalies as well as the existence of many different types of the KPIs data curves. In this paper, we propose a new anomaly detection model based on mining six local data features as the input of back-propagation (BP) neural network. By means of vectorization description on a normalized dataset innovatively, the local geometric characteristics of one time series curve could be well described in a precise mathematical way. Differing from some traditional statistics data characteristics describing the entire variation situation of one sequence, the six mined local data features give a subtle insight of local dynamics by describing the local monotonicity, the local convexity/concavity, the local inflection property and peaks distribution of one KPI time series. In order to demonstrate the validity of the proposed model, we applied our method on 14 classical KPIs time series datasets. Numerical results show that the new given scheme achieves an average F1-score over 90%. Comparison results show that the proposed model detects the anomaly more precisely.