Cartoon-Texture decomposition with patch-wise decorrelation

Point cloud completion recovers the complete point clouds from partial ones, providing numerous point cloud information for downstream tasks such as 3-D reconstruction and target detection. However, previous methods usually suffer from unstructured prediction of points in local regions and the discrete nature of the point cloud. To resolve these problems, we propose a point cloud completion network called TPDC. Representing the point cloud as a set of unordered features of points with local geometric information, we devise a Triangular Pyramid Extractor (TPE), using the simplest 3-D structure-a triangular pyramid-to convert the point cloud to a sequence of local geometric information. Our insight of revealing local geometric information in a complex environment is to design a Divide-and-Conquer Splitting Module in a Divide-and-Conquer Splitting Decoder (DCSD) to learn point-splitting patterns that can fit local regions the best. This module employs the Divide-and-Conquer approach to parallelly handle tasks related to fitting ground-truth values to base points and predicting the displacement of split points. This approach aims to make the base points align more closely with the ground-truth values while also forecasting the displacement of split points relative to the base points. Furthermore, we propose a more realistic and challenging benchmark, ShapeNetMask, with more random point cloud input, more complex random item occlusion, and more realistic random environmental perturbations. The results show that our method outperforms both widely used benchmarks as well as the new benchmark.

Dimensionality reduction via preserving local information

Edge-preserving smoothing filters have had many applications in the image processing community, such as image compression, restoration, deblurring and abstraction. However, their potential application in computer vision and machine learning has never been fully studied. The most successful feature descriptors for image classification use gradient images for extracting the overall shapes of objects, thus edge preserving filters could improve their quality. The effects of various edge-preserving filters were evaluated as a pre-processing step inhuman detection. In this work, three smoothing filters were tested, namely the total variation (TV), relative total variation (RTV) and L0 smoothing. Significant performance gains were realised with TV and RTV for both colour and thermal images while the L0 smoothing filter only realised a slight improvement on thermal images and poorer performance on colour images. These results show that smoothing filters have a potential to improve the robustness of common statistical learning classifiers.

One of the most fundamental problems in image processing and computer vision is the inherent ambiguity that exists between texture edges and object boundaries in real-world images and video. Despite this ambiguity, many applications in computer vision and image processing often use image edge strength with the assumption that these edges approximate object depth boundaries. However, this assumption is often invalidated by real world data, and this discrepancy is a significant limitation in many of today's image processing methods. We address this issue by introducing a simple, low-level, and patch-consistency assumption that leverages the extra information present in video data to resolve this ambiguity. Through analyzing how well patches can be modeled by simple transformations over time, we can obtain an indication of which image edges correspond to texture edges versus object boundaries. Our approach is simple to implement and has the potential to improve a wide range of image and video-based applications by suppressing the detrimental effects of strong texture edges on regularization terms. We validate our approach by presenting results on a variety of scene types and directly incorporating our augmented edge map into existing image segmentation and optical flow applications, showing results that better correspond to object boundaries.

A locally weighted sparse graph regularized Non-Negative Matrix Factorization method

Anomaly detection (AD) aims to distinguish the data points that are inconsistent with the overall pattern of the data. Recently, unsupervised anomaly detection methods have aroused huge attention. Among these methods, feature representation (FR) plays an important role, which can directly affect the performance of anomaly detection. Sparse representation (SR) can be regarded as one of matrix factorization (MF) methods, which is a powerful tool for FR. However, there are some limitations in the original SR. On the one hand, it just learns the shallow feature representations, which leads to the poor performance for anomaly detection. On the other hand, the local geometry structure information of data is ignored. To address these shortcomings, a graph regularized deep sparse representation (GRDSR) approach is proposed for unsupervised anomaly detection in this work. In GRDSR, a deep representation framework is first designed by extending the single layer MF to a multilayer MF for extracting hierarchical structure from the original data. Next, a graph regularization term is introduced to capture the intrinsic local geometric structure information of the original data during the process of FR, making the deep features preserve the neighborhood relationship well. Then, a L1-norm-based sparsity constraint is added to enhance the discriminant ability of the deep features. Finally, a reconstruction error is applied to distinguish anomalies. In order to demonstrate the effectiveness of the proposed approach, we conduct extensive experiments on ten datasets. Compared with the state-of-the-art methods, the proposed approach can achieve the best performance.

We propose a new regularization scheme for stable numerical differentiation of noisy data defined on a bounded domain with . The method generates a sequence of smoothed (regularized) data obtained by solving a perturbed elliptic boundary-value problem. Assuming the measured data are in and the true underlying data are sufficiently smooth, we prove convergence results in the -norm, provided the noisy data converge to true data in the sense. Using the finite element method, we derive error bounds and prove convergence theorems in the case of discrete data. Numerical examples indicate noteworthy results and shed light on some possible applications in image processing and computer vision.

Effective Image Segmentation with Flexible ICM-Based Markov Random Fields in Distributed Systems of Personal Computers

A review is presented of the use of hierarchical data structures for the representation of shape. The presentation is primarily in terms of variants of a spatial data structure known as a quadtree. These data structures are based on a recursive decomposition of the interior of the shape or on its boundary. The focus is on the representation of two-dimensional shape used in applications in image processing and computer vision. The adaptation of the concepts of a distance, a skeleton, and a medial axis transform to a shape represented by a quadtree is described, as is a region expansion (i.e., image dilation) algorithm that executes in time independent of the radius of expansion.

Describes the design and implementation of a two-dimensional systolic array processor for applications in image processing and computer vision. The processor architecture is based on a SIMD array of 4-bit processing elements, interconnected by a mesh network with four nearest neighbors. The PE array is programmable allowing the user to develop application-specific algorithms for performing analysis on image data. A prototype VLSI chip has been designed implementing a single PE and has been submitted for fabrication. The chip is expected to operate at 25 MHz. >

Image mosaicing is the act of combining two or more images and is used in many applications in computer vision, image processing, and computer graphics. It aims to combine images such that no obstructive boundaries exist around overlapped regions and to create a mosaic image that exhibits as little distortion as possible from the original images. Most of the existing algorithms are the computationally complex and don't show good results always in obtaining of the stitched images, which are different: scale, light, various free points of view and others. In this paper we consider an algorithm which allows increasing the speed of processing in the case of stitching high-resolution images. We reduced the computational complexity used an edge image analysis and saliency map on high-detailisation areas. On detected areas are determined angles of rotation, scaling factors, the coefficients of the color correction and transformation matrix. We define key points using SURF detector and ignore false correspondences based on correlation analysis. The proposed algorithm allows to combine images from free points of view with the different color balances, time shutter and scale. We perform a comparative study and show that statistically, the new algorithm deliver good quality results compared to existing algorithms.

Curve smoothing has two important applications in computer vision and image processing: 1) the curvature scale-space (CSS) technique for shape analysis, and 2) the Gaussian filter for noise suppression. In this paper, we study how planar curves converge as they are smoothed with increasing scales. First, two types of convergence behavior are clarified. The coined term shrinkage refers to the reduction of arc-length of a smoothed planar curve, which describes the convergence of the curve latitudinally; and another coined term collapse refers to the movement of each point to its limiting position, which describes the convergence of the curve longitudinally. A systematic study on the shrinkage and collapse of three categories of curve models is then presented. The corner models helps to reveal how the local structures of planar curves collapse and what the smoothed curves may converge to. The sawtooth models allows us to gain insights regarding how noise is suppressed from noisy planar curves by the Gaussian filter. Our investigation on the closed curves shows that each curve collapses to a point at its center of mass. However, different curves may yield different limiting shapes at the infinity scale. Finally, based upon the derived results the performance of the CSS technique in corner detection and shape representation is analyzed, and a fast implementation method of the Gaussian filter for noise suppression is proposed.

Tracking is recently formulated as a problem of discriminating the object from its nearby background, where the classifier is updated by new samples successively arriving during tracking. Depending on whether labeling the samples or not, the tracker can be designed in a supervised or semi-supervised manner. This paper proposes a novel semi-supervised algorithm for tracking by combining Semi-supervised Discriminant Analysis (SDA) with an online boosting framework. Using the local geometric structure information from the samples, the SDA-based weak classifier is made more robust to outliers. Meanwhile, we design an incremental updating mechanism for SDA so that it can adapt to appearance changes. We further propose an Extended SDA (ESDA) algorithm, which gives better discrimination ability. Results on several challenging video sequences demonstrate the effectiveness of the method.

Electrochemical performance of manganese hexacyanoferrate cathode material in aqueous Zn-ion battery

For classifying images with various appearances, graph embedding based subspace learning has difficulty in taking a comprehensive consideration of both local geometrical structure and between-class discriminative information. In addition, when no sufficient training samples exist, using only the simple weight graph corresponding to labeled samples, the embedding subspace may not be accurately modeled. We present a semisupervised graph embedding algorithm by combining graph embedding and sparse representation. This algorithm can effectively learn a compact and semantic subspace by using a locally connected graph, which can model the geometrical structure and essential correlation of subclusters within a class and can fully utilize both labeled and unlabeled samples. Moreover, using L2,1-norm, the proposed algorithm can preserve the sparse representation property of images from the original space in the lower dimensional projected space. Our experiments demonstrate that the proposed algorithm has better performance than the alternatives reported in recent literature.