Texture Generation Research Articles

Uncertainty-Aware Semantic Guidance and Estimation for Image Inpainting

Completing a corrupted image by filling in correct structures and reasonable textures for a complex scene remains an elusive challenge. In case that a missing hole involves diverse semantic information, conventional two-stage approaches based on structural information often lead to unreliable structural prediction and ambiguous visual texture generation. To address the problem, we propose a SEmantic GUidance and Estimation Network (SeGuE-Net) that iteratively evaluates the uncertainty of inpainted visual contents based on pixel-wise semantic inference and optimize structural priors and inpainted contents alternatively. Specifically, SeGuE-Net utilizes semantic segmentation maps as guidance in each iteration of image inpainting, under which location-dependent inferences are re-estimated, and, accordingly, poorly-inferred regions are refined in subsequent iterations. Extensive experiments on real-world images demonstrate the superiority of our proposed method over state-of-the-art approaches in terms of clear boundaries and photo-realistic textures.

Missing data reconstruction in VHR images based on progressive structure prediction and texture generation

Very high resolution (VHR) satellite and aerial images often suffer from scene occlusion caused by redundant objects. The task of removing these redundant objects can be solved by missing data reconstruction technology. However, when dealing with VHR images with large-scale missing regions, existing spatial-based methods often destroy the structural information of ground objects. To alleviate this problem, this paper proposes a novel missing data reconstruction method based on deep learning. The reconstruction process is divided into two parts: structure prediction and texture generation. First, a progressive edge generation network (PEGN) is designed to predict the edges of objects in missing regions in a progressive manner. Then, the edge map predicted by PEGN is input to a texture generation network (TGN) as structural information to produce the reconstruction results. This is a spatial-based method that can produce realistic and reasonable results without any need for auxiliary spectral or temporal data. Experiments demonstrate that our model can better restore the structure of ground objects in VHR images than other spatial-based methods and outperform them in SSIM and PSNR indices. In addition, our model also has a strong generalization capability by introducing Poisson blending and histogram matching.

3D reconstruction of human head based on consumer RGB-D sensors

Three-dimensional (3D) reconstruction of a human head with high precision has promising applications in scientific research, product design and other fields. However, it still faces resistance from two factors. One is inaccurate registration caused by symmetrical distribution of head feature points, and the other is economic burden due to high-accuracy sensors. Research on 3D reconstruction with portable consumer RGB-D sensors such as the Microsoft Kinect has been highlighted in recent years. Based on our multi-Kinect system, a precise and low-cost three-dimensional modeling method and its system implementation are introduced in this paper. A registration method for multi-source point clouds is provided, which can reduce the fusion differences and reconstruct the head model accurately. In addition, a template-based texture generation algorithm is presented to generate a fine texture. The comparison and analysis of our experiments show that our method can reconstruct a head model in an acceptable time with less memory and better effect.

Characterization of Surface Topographies Generated using Circular- and Cylindrical-Face EDT

Large area texturing using EDT is challenging due to surface damage resulted from inefficient discharges, such as arcs and short-circuits. Also, three-dimensional analyses of surface topography generated during electrical discharge texturing (EDT) are scarcely performed in the available literature. Therefore, in this work, two modes of EDT were developed for large area texturing based on the mode of electrode movement resulting in texture generation on work surfaces: circular-face EDT (CirEDT) and cylindrical-face EDT (CylEDT), and the generated surface topographies are characterized in terms of topography images, areal texture parameters, and scanning electron microscopy images. Parametric analyses are also performed to study the influences of the operating parameters, such as discharge current, pulse on-time, and peripheral velocity, on different areal texture parameters. Topography analyses reveal that these surfaces have a characteristic stochastic distribution of microstructures in shape, size, and location. Height parameter analyses reveal that CylEDT surfaces are twice positively skewed and have a higher kurtosis as compared to the CirEDT surfaces. Spatial parameter analyses reveal that both surfaces display dominance of isotropic nature and non-directional structures. CirEDT surfaces have higher lubricant retention capacity, but lesser wear volume available for running-in period as compared to CylEDT surfaces, as per volume parameter analyses. Discharge current and pulse on-time are identified as the significant parameters influencing the majority of the areal texture parameters analyzed. SEM analysis revealed that CylEDT results in a textured surface with sea-wave-like structures, whereas CirEDT results in a textured surface with lunar-craters-like structures.

Microstructure and Texture Evolution of Mg-Gd-Y-Zr Alloy during Reciprocating Upsetting-Extrusion.

Reciprocating Upsetting-Extrusion (RUE) deformation process can significantly refine the grains size and weaken the basal plane texture by applying a large cumulative strain to the alloy, which is of great significance to weaken the anisotropy of magnesium (Mg) alloys and increase the application range. In this paper, the Mg-8.27Gd-3.18Y-0.43Zr (wt %) alloy was subjected to isothermal multi-passes RUE. The microstructure and texture evolution, crystal orientation-dependent deformation mechanism of the alloy after deformation were investigated. The results clearly show that with the increase of RUE process, the grains are significantly refined through continuous dynamic recrystallization (CDRX) and discontinuous dynamic recrystallization (DDRX) mechanisms, the uniformity of the microstructure is improved, and the texture intensity is reduced. At the same time, a large number of particle phases are dynamically precipitated during the deformation process, promoting grain refinement by the particle-stimulated nucleation (PSN) mechanism. The typical [10-10] fiber texture is produced after one pass due to the basal plane of the deformed grains with a relatively high proportion is gradually parallel to the ED during extrusion process. However, the texture concentration is reduced compared with the traditional extrusion deformation, indicating that the upsetting deformation has a certain delay effect on the subsequent extrusion texture generation. After three or four passes deformation, the grain orientation is randomized due to the continuous progress of the dynamic recrystallization process.

Surface modification for osseointegration of Ti6Al4V ELI using powder mixed sinking EDM

Biomedical implant rejection due to micromotion and inflammation around an implant leads to osteolysis and eventually has an implant failure because of poor osseointegration. To enhance osseointegration, the implant surface modification both at the nano and micro-scale levels is preferred to result in an enhanced interface between the body tissue and implant. The present study focuses on the modification of the surface of Titanium (α+β) ELI medical grade alloy using powder-mixed electric discharge machining (PMEDM). Pulse current, on/off time, and various silicon carbide (SiC) powder concentrations are used as input parameters to comprehend desired surface modifications. Powder concentration is considered as the most important factor to control surface roughness and recast layer depth. A significant decrease in surface fracture density and roughness is observed using a 20 g/l concentration of SiC particles. Elemental mapping analysis has confirmed the migration of Si and the generation of promising surface texture and chemistry. Oxides and carbides enriched surface improved the microhardness of the re-solidified layer from 320 HV to 727 HV. Surface topology reveals nano-porosity (50–200 nm) which enhances osseointegration due to the absorption of proteins especially collagen to the surface.

Visco-plastic self-consistent modeling of crystallographic texture evolution related to slip systems activated during machining Ti-6AL-4V

Exploring the interconnection between texture variation and the activations of slip systems is vital for the texture generation and controlling during machining of Ti-6AL-4V. The evolution of crystallographic texture and the activations of different slip systems during machining of Ti-6AL-4V were studied combined with the orientation imaging microscopy (OIM) and visco-plastic self-consistent (VPSC) framework. A VPSC model considering the activations of five slip families as well as twinning of hexagonal close packed (HCP) Ti structure and coupled with a finite element cutting simulation model were used to simulate the texture variations during machining of Ti-6AL-4V. Crystallographic texture evolutions of Ti-6AL-4V in chips and machined surface were examined using OIM. Two initial textures of Ti-6AL-4V yielded the same end texture, a strong C fiber texture. Lastly, Voce hardening parameters were modified to control the activations of slip modes as well as twinning and calibrate the VPSC simulation results of texture and mechanical property. It is found that the combination of prismatic <a>, basal <a>, pyramidal <a> and 2-nd order pyramidal <c+a> slip systems induce the generation of C shear texture in machining of Ti-6AL-4V. The activation of 2-nd order <c+a> slip system makes large contribution to the formation of C shear texture irrespective of the initial textures and machining parameters. The result demonstrates that activations of slip systems and their relative activities depend on the shear deformation in machining of Ti-6AL-4V. This finding is important to control the texture variation in the deformation of Ti-6AL-4V.

Optimization of UAVs‐SfM data collection in aeolian landform morphodynamics: a case study from the Gonghe Basin, China

AbstractUAVs‐SfM (unmanned aerial vehicles‐structure‐from‐motion) systems can generate high‐resolution three‐dimensional (3D) topographic models of aeolian landforms. To explore the optimization of UAVs‐SfM for use in aeolian landform morphodynamics, this study tested flight parameters for two contrasting aeolian landform areas (free dune and blowout) to assess the 3D reconstruction accuracy of the UAV survey compared with field point measurements using differential RTK‐GPS (real‐time kinematic‐global positioning system). The results reveal the optimum UAVs‐SfM flight set‐up at the free‐dune site was: flying height = 74 m, camera tilt angle = −90°, photograph overlap ratio = 85%/70% (heading/sideways). The horizontal/vertical location error was around 0.028–0.055 m and 0.053–0.069 m, respectively, and a point cloud density of 463/m3 was found to generate a clear texture using these flying parameters. For the &lt; 20 m deep blowout the optimum set‐up with highest accuracy and the lowest cliff texture distortion was: flying height = 74 m combined camera tilt angle = −90° and −60°, photograph overlap ratio = 85%/70% (heading/sideways), and an evenly distributed GCPs (ground control points) density of 42/km2 using these flying parameters. When the depth of the blowouts exceeded 40 m, the optimum flight/survey parameters changed slightly to account for more challenging cliff texture generation: flying height = 80 m (with −90° and −60°combined camera tilt angle), GCPs density = 63/km2 to generate horizontal and vertical location error of 0.024 m and 0.050 m, respectively, and point cloud density of 2597.11/m3. The main external factors that affect the successful 3D reconstruction of aeolian landforms using UAVs‐SfM are the weather conditions, manipulation errors, and instrument system errors. The UAVs‐SfM topographic monitoring results demonstrate that UAVs provide a viable and robust means for aeolian landform morphodynamics monitoring. Importantly, the rapid and high precision 3D reconstruction processes were significantly advanced using the optimal flight parameters reported here. © 2020 John Wiley &amp; Sons, Ltd.

A Novel Approach to Structure Plant-Based Yogurts Using High Pressure Processing.

Current plant-based yogurts are made by the fermentation of plant-based milks. Although this imparts fermented flavors and probiotic cultures, the process is relatively longer and often leads to textural issues. The protein content of these plant-based yogurts is also lower than their dairy counterparts. To overcome these challenges, this paper explores the high pressure processing (HPP) of plant protein ingredients as an alternative structuring strategy for plant-based yogurts. Using mung bean (MB), chickpea (CP), pea (PP), lentil (LP), and faba bean (FB) proteins as examples, this work compared the viscosity and viscoelastic properties of high pressure-structured (600 MPa, 5 min, 5 °C) 12% (w/w) plant protein gels without, and with 5% (w/w) sunflower oil (SO) to commercial plain skim and whole milk Greek yogurts and discussed the feasibility of using HPP to develop plant-based yogurts. HPP formed viscoelastic gels (G’ > G’’) for all plant protein samples with comparable gel strength (G’~102–103 Pa; tan δ~0.2–0.3) to commercial dairy yogurts. The plant protein gel strength decreased in the order: CP~CPSO~LP~LPSO > MBSO~PPSO~FB~FBSO > PP >> MB. Modest addition of sunflower oil led to little change in viscoelastic properties for all plant protein samples except for MB and PP, where gel strength increased with incorporated oil. The emulsion gels were also more viscous than the hydrogels. Nonetheless, the viscosity of the plant protein gels was similar to the dairy yogurts. Finally, a process involving separate biotransformation for optimized flavor production and high pressure processing for consistent texture generation was proposed. This could lead to high protein plant-based yogurt products with desirable texture, flavor, and nutrition.

A New Volumetric Fusion Strategy with Adaptive Weight Field for RGB-D Reconstruction.

High-quality 3D reconstruction results are very important in many application fields. However, current texture generation methods based on point sampling and fusion often produce blur. To solve this problem, we propose a new volumetric fusion strategy which can be embedded in the current online and offline reconstruction framework as a basic module to achieve excellent geometry and texture effects. The improvement comes from two aspects. Firstly, we establish an adaptive weight field to evaluate and adjust the reliability of data from RGB-D images by using a probabilistic and heuristic method. By using this adaptive weight field to guide the voxel fusion process, we can effectively preserve the local texture structure of the mesh, avoid wrong texture problems and suppress the influence of outlier noise on the geometric surface. Secondly, we use a new texture fusion strategy that combines replacement, integration, and fixedness operations to fuse and update voxel texture to reduce blur. Experimental results demonstrate that compared with the classical KinectFusion, our approach can significantly improve the accuracy in geometry and texture clarity, and can achieve equivalent texture reconstruction effects in real-time as the offline reconstruction methods such as intrinsic3d, even better in relief scenes.

Texture evolution in nanocrystalline Cu under shock compression

We investigate with large-scale molecular dynamics simulations texture evolution in nanocrystalline Cu under planar shock wave loading. Five representative initial textures are explored under varying impact strengths. On the basis of Euler angles, we perform orientation mapping and texture analysis, including orientation distribution functions, pole figures, and inverse pole figures. Shock compression induces a weak but apparent ⟨110⟩ textures in nanocrystalline Cu initially with no texture, and a ⟨100⟩ fiber texture, and an incomplete weak ⟨110⟩ texture in nanocrystalline Cu initially with a {100}⟨100⟩ recrystallization texture; such texture changes can be attributed to deformation twinning and dislocation slip and traced back to grains initially with ⟨100⟩. A ⟨100⟩ texture and a {100}⟨100⟩ cube texture component are induced via twinning in nanocrystalline Cu initially with the ⟨111⟩ and β fiber textures, respectively, and can be traced back to grains initially with ⟨111⟩.

Hollow carbon microbox from acetylacetone as anode material for sodium-ion batteries

Abstract Carbon-based materials have attracted much interest as one of the promising anodes for sodium-ion batteries. However, low utilization of electrolyte and slow ion-transfer rate during electrochemical process hinder the further application of traditional bulk carbon. In order to enhance the diffusion kinetics and maintain the reversibility, hierarchical hollow carbon microbox was successfully prepared through a tunable bottom-up self-template routine for sodium-ion batteries. During annealing process, the morphology construction and activation happened synchronously. Based on that, a range of cross-linked porous nanosheet and hollow microbox were attained by manipulating reactant condition. The generation of texture and physical property are analyzed and are established linkages related to the electrochemical behavior. As results depicted in kinetic exploration and simulation based on cyclic voltammetry, the surface-controlled electrochemical behavior gradually turns to be the diffusion-controlled behavior as the hollow microbox evolves to porous nanosheet. The probable reason is that the rational microstructure/texture design leads to the accelerated diffusion kinetic procedure and the reduced concentration difference polarization. Sodium storage mechanism was deduced as reversible binding of Na-ions with local defects, including vacancies on sp2 graphitic layers, at the edges of flakes and other structural defects instead of intercalation. Bestowed by the morphology design, the broad pore width distribution, abundant defects/active sites and surface functionality, hollow microbox electrode delivers great electrochemical performances. This work is expected to propose a novel and effective strategy to prepare tunable hierarchical hollow carbon microbox and induce the fast kinetic of carbon anode material.

Semantics-Aligned Representation Learning for Person Re-Identification

Person re-identification (reID) aims to match person images to retrieve the ones with the same identity. This is a challenging task, as the images to be matched are generally semantically misaligned due to the diversity of human poses and capture viewpoints, incompleteness of the visible bodies (due to occlusion), etc. In this paper, we propose a framework that drives the reID network to learn semantics-aligned feature representation through delicate supervision designs. Specifically, we build a Semantics Aligning Network (SAN) which consists of a base network as encoder (SA-Enc) for re-ID, and a decoder (SA-Dec) for reconstructing/regressing the densely semantics aligned full texture image. We jointly train the SAN under the supervisions of person re-identification and aligned texture generation. Moreover, at the decoder, besides the reconstruction loss, we add Triplet ReID constraints over the feature maps as the perceptual losses. The decoder is discarded in the inference and thus our scheme is computationally efficient. Ablation studies demonstrate the effectiveness of our design. We achieve the state-of-the-art performances on the benchmark datasets CUHK03, Market1501, MSMT17, and the partial person reID dataset Partial REID.

Modelling environment for analyzing the algorithms for 3D reconstruction of videoendoscopic research objects

Three-dimensional reconstruction based on the results of video endoscopic examination is a promising area for supporting medical diagnostics and treatment planning for a wide range of pathologies. Nevertheless, the assessment of the results of such reconstruction and verification of the correspondence of the obtained three-dimensional model to the original scene is significantly challenging. As a solution to this problem, the possibility of using a modelling environment to emulate the process of obtaining source video endoscopic data from the generated scene is suggested. The problem of three-dimensional modelling of the esophagus using the Autodesk 3ds Max environment and the Arnold visualization engine is considered. The paper describes the procedural generation of textures for the model and proposes the using Periodic Spatial Generative Adversarial Network models based on convolutional neural networks. To compare the result of reconstruction with a scene, generated using the proposed modelling environment, an optimality criterion is introduced, by which the individual stages of the three-dimensional reconstruction algorithm are compared when the model is optimized using the bundle adjustment method.

Performance Assessment of Texture Reproduction in High-Resolution CT.

Assessment of computed tomography (CT) images can be complex due to a number of dependencies that affect system performance. In particular, it is well-known that noise in CT is object-dependent. Such object-dependence can be more pronounced and extend to resolution and image textures with the increasing adoption of model-based reconstruction and processing with machine learning methods. Moreover, such processing is often inherently nonlinear complicating assessments with simple measures of spatial resolution, etc. Similarly, recent advances in CT system design have attempted to improve fine resolution details - e.g., with newer detectors, smaller focal spots, etc. Recognizing these trends, there is a greater need for imaging assessment that are considering specific features of interest that can be placed within an anthropomorphic phantom for realistic emulation and evaluation. In this work, we devise a methodology for 3D-printing phantom inserts using procedural texture generation for evaluation of performance of high-resolution CT systems. Accurate representations of texture have previously been a hindrance to adoption of processing methods like model-based reconstruction, and texture serves as an important diagnostic feature (e.g. heterogeneity of lesions is a marker for malignancy). We consider the ability of different systems to reproduce various textures (as a function of the intrinsic feature sizes of the texture), comparing microCT, cone-beam CT, and diagnostic CT using normal- and high-resolution modes. We expect that this general methodology will provide a pathway for repeatable and robust assessments of different imaging systems and processing methods.

Survey of Procedural Methods for Two-Dimensional Texture Generation

Textures are the most important element for simulating real-world scenes and providing realistic and immersive sensations in many applications. Procedural textures can simulate a broad variety of surface textures, which is helpful for the design and development of new sensors. Procedural texture generation is the process of creating textures using mathematical models. The input to these models can be a set of parameters, random values generated by noise functions, or existing texture images, which may be further processed or combined to generate new textures. Many methods for procedural texture generation have been proposed, but there has been no comprehensive survey or comparison of them yet. In this paper, we present a review of different procedural texture generation methods, according to the characteristics of the generated textures. We divide the different generation methods into two categories: structured texture and unstructured texture generation methods. Example textures are generated using these methods with varying parameter values. Furthermore, we survey post-processing methods based on the filtering and combination of different generation models. We also present a taxonomy of different models, according to the mathematical functions and texture samples they can produce. Finally, a psychophysical experiment is designed to identify the perceptual features of the example textures. Finally, an analysis of the results illustrates the strengths and weaknesses of these methods.

Real-time computer vision system for tracking simultaneously subject-specific rigid head and non-rigid facial mimic movements using a contactless sensor and system of systems approach

Background and ObjectiveHead and facial mimic animations play important roles in various fields such as human-machine interactions, internet communications, multimedia applications, and facial mimic analysis. Numerous studies have been trying to simulate these animations. However, they hardly achieved all requirements of full rigid head and non-rigid facial mimic animations in a subject-specific manner with real-time framerates. Consequently, this present study aimed to develop a real-time computer vision system for tracking simultaneously rigid head and non-rigid facial mimic movements. MethodsOur system was developed using the system of systems approach. A data acquisition sub-system was implemented using a contactless Kinect sensor. A subject-specific model generation sub-system was designed to create the geometrical model from the Kinect sensor without texture information. A subject-specific texture generation sub-system was designed for enhancing the reality of the generated model with texture information. A head animation sub-system with graphical user interfaces was also developed. Model accuracy and system performances were analyzed. ResultsThe comparison with MRI-based model shows a very good accuracy level (distance deviation of ~1 mm in neutral position and an error range of [2–3 mm] for different facial mimic positions) for the generated model from our system. Moreover, the system speed can be optimized to reach a high framerate (up to 60 fps) during different head and facial mimic animations. ConclusionsThis study presents a novel computer vision system for tracking simultaneously subject-specific rigid head and non-rigid facial mimic movements in real time. In perspectives, serious game technology will be integrated into this system towards a full computer-aided decision support system for facial rehabilitation.

The Effect of Material Surface Roughness in Aluminum Forming

In the last few years many efforts have been carried out in order to understand better what the real contact between material and tools is. Based on the better understanding new friction models have been developed which have allowed process designers to improve numerical results in terms of component viability and geometrical accuracy. Aluminum forming is especially critical since aluminum alloys have a higher tendency to generate galling in the tooling. In order to avoid galling problems, several strategies have been applied such as improving the surface quality of the tools, the application of special lubricants, the coating of the tooling surface or the generation of special textures in the tooling surface. However, not many researchers have focused in the effect of the surface texture of the sheet in the final tribological conditions at the interface. In the present paper 1050 aluminum alloy has been textured under different conditions in order to generate a wide variety of surface roughness. After that, strip drawing test have been carried out and the coefficient of friction depending on the contact pressure has been calculated. Finally, an industrial component has been simulated using the coefficient of friction calculated for the different surface roughnesses. This way, the impact of the material roughness in the component viability as well as in the component geometry has been evaluated.

Full Face-and-Head 3D Model With Photorealistic Texture

In the recent period, significant progress has been achieved towards reconstructing the 3D face model from face image. With the support of the render engines and sufficient data, the reconstruction results are fine in detail. Nevertheless, the research on the 3D face reconstruction with texture from a single unrestricted face image is imperfect. The rebuild process lacks essential structure and texture information in the profile and the craniofacial region. To address this problem, we present a method of creating a 3D full face-and-head model with photorealistic texture from a single “in-the-wild” face image in this paper. To this end, we introduce a pipeline to integrate the highly-detailed face model into the basic model. Specifically, the basic model was built by multilinear optimization, and the highly-detailed face model which represents the facial features generated by constrained illumination distribution. Additionally, to infer the invisible region texture information corresponding to the input face image, we design an effective architecture with the generative adversarial network (GAN) for panoramic UV texture generation. The final results after UV texture mapping were visualized in the experiment, which demonstrates that the model faithfully recovers the photorealistic details in arbitrary perspective. Furthermore, compared to the state-of-the-art facial modeling techniques and existing commercial solutions, our method takes less input and performs better in surface detail.

Fast Texture Synthesis for Discrete Example-Based Elements

Considering the problem of discrete texture synthesis and the time for texturing, this paper proposes a novel framework for synthesizing texture images based on discrete example-based elements. We start with extracting texture feature distribution from exemplars and then produce discrete elements based on the cluster algorithm. After initializing a texture image, we propose a texture optimization algorithm based on heuristic searching to improve the quality of the texture image. Final, we use a texture transfer method based on Convolutional Neural Network (CNN) to stylize the optimized texture image. Our results show that the proposed texture synthesis method can significantly improve the quality of discrete texture synthesis and effectively shorten the time for texture generation.