Discrete Transform Research Articles

Continuous reduction and phase transformation mechanism of pellets in lumpy zone based on dissected hydrogen blast furnace

The continuous phase transition phenomena and mechanism of pellets in the hydrogen blast furnace is achieved by dissecting the 40m3 hydrogen blast furnace. The results of the dissection reveal that the lumpy zone extends to the bosh of the hydrogen blast furnace (HBF), while it ends at about 2/3 of the stack in traditional blast furnace. The iron oxide (FeOx) began to appear in the upper stack, while the iron started to appear in the lower stack. The reduction rate and metallization rate at the bosh are measured at 95.18% and 95.54%. The content of ferrous oxide (FeO) decreases significantly (measured at 1.73% at the bosh), resulting in an increase in the melting point of the primary slag and a deterioration of its fluidity. FeOx formed in the upper stack continuously diffuses and merges into the gangue, which affects the subsequent transformation of FeOx. FeOx shows a preference for interacting with SiO2, CaO, and MgO, directly influencing the formation of slag in the cohesive zone. These findings offer valuable insights for numerical simulation and research on hydrogen-rich operation.

CWT-ViT: A time–frequency representation and vision transformer-based framework for automated robotic surgical skill assessment

Surgical skill assessment currently hinges on the manual observations of senior surgeons, and the assessment process is inherently time-consuming and subjective. Hence, there is a need to develop machine learning-based automated robotic surgical skill assessment. However, the existing machine learning-based works are only built in either the time domain or frequency domain but have never considered the investigation on the time–frequency domain. To fill the research gap, we explore the representation of the surgery motion data in the time–frequency domain. In this study, we propose a novel automated robotic surgical skill assessment framework called Continuous Wavelet Transform-Vision Transformer (CWT-ViT). We apply continuous wavelet transform, i.e., a time–frequency representation method, to convert robotic surgery kinematic data to synthesis images. Furthermore, by taking advantage of the prior knowledge of the da Vinci surgical system, we design a four branches-based architecture, each branch representing a robotic manipulator. We have conducted extensive experiments and achieved comparable results on the benchmark robotic surgical skill dataset JHU-ISI Gesture and Skill Assessment Working Set (JIGSAWS). Our proposed CWT-ViT framework has demonstrated the feasibility of applying time–frequency representation for automated robotic surgical skill assessment using kinematic data. The code is available at https://github.com/yiming95/CWT-ViT-Surgery.

Relative kinetic stability of defect patterns in two-dimensional nematic liquid crystals with rectangular confinement.

Guiding and dynamically modulating topological defects are critical challenges in defect engineering of liquid crystals. Here, we employ molecular dynamics simulations to investigate the transition dynamics and relative kinetic stability of defect patterns in two-dimensional nematic Gay-Berne liquid crystals confined within rectangular geometries. We observe the formation of various defect patterns including long-axis, diagonal, X-shaped, composite, and bend configurations under different confinement conditions. The competition between boundary effects and the uniformity of nematic orientation induces the continuous realignment of liquid crystal molecules, facilitating the spatially continuous transformation of defect patterns over time. This transition involves changes in both defect types and their locations, typically initiating from defect regions. Furthermore, we demonstrate that the relative stability of these defect patterns can be effectively controlled by adjusting confinement parameters and external field conditions. Our findings provide fundamental insights into the transition kinetics of defect patterns in confined nematic liquid crystals, thereby enhancing our ability to manipulate topological defects for advanced applications.

Exploring stability of Jeffrey fluids in anisotropic porous media: incorporating Soret effects and microbial systems

Purpose This paper aims to present a study on stability analysis of Jeffrey fluids in the presence of emergent chemical gradients within microbial systems of anisotropic porous media. Design/methodology/approach This study uses an effective method that combines non-dimensionalization, normal mode analysis and linear stability analysis to examine the stability of Jeffrey fluids in the presence of emergent chemical gradients inside microbial systems in anisotropic porous media. The study focuses on determining critical values and understanding how temperature gradients, concentration gradients and chemical reactions influence the onset of bioconvection patterns. Mathematical transformations and analytical approaches are used to investigate the system’s complicated dynamics and the interaction of numerous characteristics that influence stability. Findings The analysis is performed using the Jeffrey-Darcy type and Boussinesq estimation. The process involves using non-dimensionalization, using the normal mode approach and conducting linear stability analysis to convert the field equations into ordinary differential equations. The conventional thermal Rayleigh Darcy number RDa,c is derived as a comprehensive function of various parameters, and it remains unaffected by the bio convection Lewis number Łe. Indeed, elevating the values of ζ and γ′ in the interval of 0 to 1 has been noted to expedite the formation of bioconvection patterns while concurrently expanding the dimensions of convective cells. The purpose of this investigation is to learn how the temperature gradient affects the concentration gradient and, in turn, the stability and initiation of bioconvection by taking the Soret effect into the equation. The results provide insightful understandings of the intricate dynamics of fluid systems affected by chemical and biological elements, providing possibilities for possible industrial and biological process applications. The findings illustrate that augmenting both microbe concentration and the bioconvection Péclet number results in an unstable system. In this study, the experimental Rayleigh number RDa,c was determined to be 4π2at the critical wave number ( δcˇ) of π. Originality/value The study’s novelty originated from its investigation of a novel and complicated system incorporating Jeffrey fluids, emergent chemical gradients and anisotropic porous media, as well as the use of mathematical and analytical approaches to explore the system’s stability and dynamics.

MCGnet with multi-level gated unit transformations: A time series prediction-based model for unknown pattern abnormality detection

The task of anomaly detection in data patterns remains challenging due to the diverse of fault patterns, which often render existing models ineffective when encountered with unknown or unseen data patterns in the training data. Furthermore, insufficient samples, the presence of chaotic behavior, unbalanced anomaly patterns are potential limiting factors in the practical application. While prediction error-based approaches attempt to address these challenges, the discriminability of abnormal patterns in existing scenarios is not yet sufficient, thus limiting the ability to identify abnormal phenomena. To overcome these limitations, we proposed a novel model called MCGnet. The backbone architecture of MCGnet employed a novel design of continuous multi-layer circular convolution transformations. This design enhances the expressive capability of the model by expanding the differences among data patterns. Additionally, we introduced in parallel the newly proposed temporal-frequency gated Gaussian unit, which generates richer feature representations. Subsequently, the obtained feature representations were fed into an explicit causal network to learn complex patterns of data along temporal dependencies. The MCGnet not only solves high-precision time series prediction problems but also significantly improves the discriminability between normal and abnormal patterns. Since the proposed model is accurately modeled using only a small amount of normal data, it bypasses the limiting factors such as insufficient fault samples and sample imbalance, and is able to effectively deal with various unknown data patterns. The accuracy of the proposed method was verified through simulation data (Duffing system) and its potential for engineering applications was demonstrated through real-world data (rotor system).

Methods for Detecting the Patient’s Pupils’ Coordinates and Head Rotation Angle for the Video Head Impulse Test (vHIT), Applicable for the Diagnosis of Vestibular Neuritis and Pre-Stroke Conditions

In the contemporary era, dizziness is a prevalent ailment among patients. It can be caused by either vestibular neuritis or a stroke. Given the lack of diagnostic utility of instrumental methods in acute isolated vertigo, the differentiation of vestibular neuritis and stroke is primarily clinical. As a part of the initial differential diagnosis, the physician focuses on the characteristics of nystagmus and the results of the video head impulse test (vHIT). Instruments for accurate vHIT are costly and are often utilized exclusively in healthcare settings. The objective of this paper is to review contemporary methodologies for accurately detecting the position of pupil centers in both eyes of a patient and for precisely extracting their coordinates. Additionally, the paper describes methods for accurately determining the head rotation angle under diverse imaging and lighting conditions. Furthermore, the suitability of these methods for vHIT is being evaluated. We assume the maximum allowable error is 0.005 radians per frame to detect pupils’ coordinates or 0.3 degrees per frame while detecting the head position. We found that for such conditions, the most suitable approaches for head posture detection are deep learning (including LSTM networks), search by template matching, linear regression of EMG sensor data, and optical fiber sensor usage. The most relevant approaches for pupil localization for our medical tasks are deep learning, geometric transformations, decision trees, and RASNAC. This study might assist in the identification of a number of approaches that can be employed in the future to construct a high-accuracy system for vHIT based on a smartphone or a home computer, with subsequent signal processing and initial diagnosis.

Research on Vibration Control Regarding Mechanical Coupling for Maglev Trains with Experimental Verification

The electromagnet module, as a fundamental component providing levitation force for maglev trains, plays a crucial role in ensuring the stability of train operation. However, vibrations can easily occur due to the mechanical coupling between the two suspension points of the electromagnet module. To reveal the inherent instability of the system and the coupling relationship between the state variables, a state-space equation that considers the mechanical coupling between the two suspension points is established. Furthermore, a differential control algorithm based on geometric feature transformation is proposed to mitigate the structural coupling vibration. Simulation experiments are conducted to compare the dynamic characteristics of the system before and after implementing the improvement algorithm under complex conditions. At the same time, the influence of control parameters on electromagnetic vibration was analyzed, focusing particularly on vibrations resulting from parameter mismatch, offering crucial insights for enhancing system stability. Additionally, suspension tests are carried out on the high-speed double bogie test platform in the Key Laboratory of Hunan Province to further validate the effectiveness of the proposed algorithm. The proposed control framework is both effective and concise, making it easy to implement in engineering applications. This research holds significant practical value in improving the stability of maglev trains.

Intraoperative detection of parathyroid glands using artificial intelligence: optimizing medical image training with data augmentation methods

BackgroundPostoperative hypoparathyroidism is a major complication of thyroidectomy, occurring when the parathyroid glands are inadvertently damaged during surgery. Although intraoperative images are rarely used to train artificial intelligence (AI) because of its complex nature, AI may be trained to intraoperatively detect parathyroid glands using various augmentation methods. The purpose of this study was to train an effective AI model to detect parathyroid glands during thyroidectomy.MethodsVideo clips of the parathyroid gland were collected during thyroid lobectomy procedures. Confirmed parathyroid images were used to train three types of datasets according to augmentation status: baseline, geometric transformation, and generative adversarial network-based image inpainting. The primary outcome was the average precision of the performance of AI in detecting parathyroid glands.Results152 Fine-needle aspiration-confirmed parathyroid gland images were acquired from 150 patients who underwent unilateral lobectomy. The average precision of the AI model in detecting parathyroid glands based on baseline data was 77%. This performance was enhanced by applying both geometric transformation and image inpainting augmentation methods, with the geometric transformation data augmentation dataset showing a higher average precision (79%) than the image inpainting model (78.6%). When this model was subjected to external validation using a completely different thyroidectomy approach, the image inpainting method was more effective (46%) than both the geometric transformation (37%) and baseline (33%) methods.ConclusionThis AI model was found to be an effective and generalizable tool in the intraoperative identification of parathyroid glands during thyroidectomy, especially when aided by appropriate augmentation methods. Additional studies comparing model performance and surgeon identification, however, are needed to assess the true clinical relevance of this AI model.Graphical abstract

Effect of Combined Addition of Molybdenum and Tungsten on Continuous Cooling Transformation Behavior of High Chromium Cast Iron

Effect of Combined Addition of Molybdenum and Tungsten on Continuous Cooling Transformation Behavior of High Chromium Cast Iron

Electrocatalytic Selenate Transformation to Tunable Products Using Affordable Catalysts

Anthropogenic selenium (Se) pollution has profound ecosystem and human health impacts. Compared to selenite or Se(IV), selenate or Se(VI) is chemically inert and more challenging for selective removal from the complex aquatic environment.1 Current Se(VI) removal efforts seek intensive resource and energy input to drive Se(VI) transformation, and are subject to environmental and operating conditions.2 The potential solution for sustainable Se(VI) transformation involves catalytic reduction of Se(VI). In current catalytic approaches, the reduction of Se(VI) is achieved using either abiotic photocatalysts (e.g., TiO2) or biotic enzymes (e.g., nitrate reductase).3 Here, we explored the opportunities of electrified Se(VI) transformation with functionalized catalysts to manage aquatic selenate pollution. Se(VI) can be removed through phase transformation into solid Se(0) that could cover reactive surfaces and prevent further catalysis, or gaseous Se(-II) that is toxic and needs to be captured. Alternatively, the reduction of Se(VI) into reactive Se(IV) allows exposure of active interfaces for continuous electrochemcial transformation. Therefore, a controlled reduction of Se(VI) to Se(IV) would be ideal in aquatic Se(VI) management and potential Se recovery. To the best of our knowledge, controlled Se(VI) transformation has not been fully achieved in the literature. In this study, we evaluated five affordable catalysts on graphite cathodes, including TiO2,3 underpotentially deposited Cu (UPD Cu),4 underpotentially deposited Cd (UPD Cd), Co, and CuFe nanoarrays.5 Among these five candidates, characteristic Se(VI) reduction peaks were identified for TiO2, UPD Cu, and UPD Cd through cyclic voltammetry. Less than 5% of 1-mM Se was removed using UPD Cu and UPD Cd, based on 24-hour chronoamperometry tests at corresponding potentials. Unfortunately, ppm-level Cu or Cd were detected in the treated effluent. In contrast, more than 90% of Se(VI) was removed using TiO2 as the catalyst after 24 hours. TiO2 was more robust and stable in electrocatalytic reactions with minimal catalyst dissolution and, therefore, was selected for further electrocatalytic Se(VI) reduction studies.During the TiO2-catalyzed reduction, Se(VI) would first transform into Se(IV) at a relatively positive potential. A maximum of 97% Se(VI) was found to be converted to Se(IV) in a 6-hour electrocatalytic Se(VI) reduction at -0.18 V. Decreasing electrode potential to -0.35 V, solid deposits of Se(0) appeared on a TiO2-modified graphite cathode. With a more negative electrode potential, further Se(0) reduction to Se(-II) was identified. The findings spotlight the opportunities to control Se(VI) transformation and harvest tunable products per removal demand and strategies. The effect of operational conditions was investigated at various TiO2 loading and solution pH. Higher TiO2 loading and lower solution pH favored the electrocatalytic Se(VI) transformation. The removal kinetics and performance were then studied under optimal operational conditions at -0.45 V, where Se(0) was dominant among Se(VI) reduction products, and -0.70 V, where Se(-II) was abundant. Kinetics analysis revealed that electrocatalytic Se(VI) reduction at -0.70 V was roughly four times faster as compared to that at -0.45 V. A total of 88.7 ± 2.3% of Se(VI) was removed after 24 hours of electrocatalysis at -0.70 V. In contrast, 48.2 ± 9.3% of Se(VI) was removed after 24 hours of electrocatalysis at -0.45 V. These results warrant further evaluation of the regenerability and lifespan of TiO2-modified graphite electrodes, an optimized electrode/reactor design, and an integrated Se removal/recovery system.

Applied for Higher Order Bezier Curve using MATLAB

A Bézier curve is a parametric curve used in computer graphics and related fields. A set of discrete "control points" defines a smooth, continuous curve by means of a formula. The piecewise nature of a Bezier curve means that its representative equation is a linear combination of Bezier of particular degrees. The aim of this study is to infer Bézier curve graphs from higher orders using MATLAB. Where the applied analytical method was used I have reached .Will be clarified, and the following results have been reached is used in the linear curve association to express the distance ( on the line between the points and , In a second-order Bezier curve, we can create intermediate points such as and . These points have a value located in the interval , To obtain a Bezier curve for higher degrees, note that the idea in arriving at drawing the curve is to divide each line by placing a point on it and then connect the new points to each other (note that we get rid of a point at each stage), and we repeat the process until we have only one line left. So, we put ( on it and then we draw the curve. By comparing the manual solution and the solution using MATLAB, we find that the solution using MATLAB is faster and more accurate, especially at higher levels. And the study recommends the following : It is noted that the curve lies completely within the convex closure of the points, so these points are used with ease to form the required curve, it is also possible to apply geometric transformations (such as rotation and sliding) to the curve by applying this transformation to the control points of this curve.

Continuous polyamorphic transition in high-entropy metallic glass

Polyamorphic transition (PT) is a compelling and pivotal physical phenomenon in the field of glass and materials science. Understanding this transition is of scientific and technological significance, as it offers an important pathway for effectively tuning the structure and property of glasses. In contrast to the PT observed in conventional metallic glasses (MGs), which typically exhibit a pronounced first-order nature, herein we report a continuous PT (CPT) without first-order characteristics in high-entropy MGs (HEMGs) upon heating. This CPT behavior is featured by the continuous structural evolution at the atomic level and an increasing chemical concentration gradient with temperature, but no abrupt reduction in volume and energy. The continuous transformation is associated with the absence of local favorable structures and chemical heterogeneity caused by the high configurational entropy, which limits the distance and frequency of atomic diffusion. As a result of the CPT, numerous glass states can be generated, which provides an opportunity to understand the nature, atomic packing, formability, and properties of MGs. Moreover, this discovery highlights the implication of configurational entropy in exploring polyamorphic glasses with an identical composition but highly tunable structures and properties.

Comparison of Fourier and Trigonometric Transform based Multicarrier Modulations for Visible Light Communication

It is predicted that the radio frequency spectrum will be insufficient in the near future due to the increase in wireless data. Visible Light Communication (VLC) is an alternative solution, which promises high speeds. Similar to other wireless communication systems, VLC systems prefer Multicarrier Modulation (MCM), but the signals are converted to be real and unipolar before transmission for optical communication. In this paper, two optical MCM groups that utilize Discrete Fourier Transform (DFT) and Discrete Trigonometric Transform (DTT) are questioned with respect to Bit Error Rate (BER), spectral efficiency and complexity. DFT based techniques use complex mapped signals together with their Hermitian symmetries to obtain real output signals, while DTT based techniques already output real signals when the input signal is real mapped. It is seen that DFT based techniques have lower BERs because of used mapping. DTT based techniques improve spectral efficiency, but they are limited to real mappings with higher error rates. For both transformations, the real signals are made unipolar by adding a bias (DCO-MCM), by asymmetrically clipping (ACO-MCM) or by sending positive and negative values separately (UnO-MCM). It is shown that, adding a dc bias (DCO-MCM) increases BERs, where ACO-MCM and UnO-MCM have close performances with lower BERs.

Quantum melting of long-range ordered quantum antiferromagnets investigated by momentum-space continuous similarity transformations

Quantum melting of long-range ordered quantum antiferromagnets investigated by momentum-space continuous similarity transformations

Evaluation of Deformable Convolution: An Investigation in Image and Video Classification

Convolutional Neural Networks (CNNs) present drawbacks for modeling geometric transformations, caused by the convolution operation’s locality. Deformable convolution (DCON) is a mechanism that solves these drawbacks and improves the robustness. In this study, we clarify the optimal way to replace the standard convolution with its deformable counterpart in a CNN model. To this end, we conducted several experiments using DCONs applied in the layers that conform a small four-layer CNN model and on the four-layers of several ResNets with depths 18, 34, 50, and 101. The models were tested in binary balanced classes with 2D and 3D data. If DCON is used on the first layers of the proposal of model, the computational resources will tend to increase and produce bigger misclassification than the standard CNN. However, if the DCON is used at the end layers, the quantity of Flops will decrease, and the classification accuracy will improve by up to 20% about the base model. Moreover, it gains robustness because it can adapt to the object of interest. Also, the best kernel size of the DCON is three. With these results, we propose a guideline and contribute to understanding the impact of DCON on the robustness of CNNs.

The Material Politics of Graffiti Removal

While anti-graffiti policies have been examined in various places, the concrete conditions of graffiti removal remain largely unexplored. Based on ethnographic fieldwork, this article considers a generative dimension of graffiti removal techniques and highlights their contribution to the visual ordering of urban reality through two main actions: making differences and making similarities. Anything but automatic and simple, graffiti removal is a complex operation consisting of technical gestures and ontological re-specifications, which are accomplished within the material entanglements of the city’s surfaces. A matter of continuous transformations, graffiti removal participates, like graffiti writing, in the visual becoming of urban ecologies.

A Critical Examination of Digital Competency among Postgraduate Students of Burdwan University

In the age of continuous transformation, keeping in view the gradual increase in the usage of digital platforms, this study aims to explore digital competency among postgraduate students of Burdwan University. In this study, researchers have used a standardized questionnaire developed by Tzaflkou et al. [1] consisting of 28 items. 5- The Point Likert scale has been used for this research. Among randomly selected 112 respondents 65 were boys and 47 were girls in this study. The researchers used the “t-test” to analyze the data. The evaluated data revealed an insignificant difference in digital competency between male and female postgraduate students, whereas a significant difference in digital competency at .05 level between postgraduate students belonging to the arts and science stream prevailed here. Between Urban and rural postgraduate students, there was a significant difference at 0.05 level. The justification of this study shows that digital competence can make students more capable, and more job-oriented than before. The recommendations regarding digital competency for students, teachers, and overall educational institutions proved that it is essential for upgrading themselves and coping with the modern world. In conclusion, the study revealed that the proposed scale could be utilized for new acts and schemes regarding lifelong learning, and planned modifications, in the solitary areas for the betterment of people including students.

Lattice Hamiltonian for adjoint QCD2

We introduce a Hamiltonian lattice model for the (1 + 1)-dimensional SU(Nc) gauge theory coupled to one adjoint Majorana fermion of mass m. The discretization of the continuum theory uses staggered Majorana fermions. We analyze the symmetries of the lattice model and find lattice analogs of the anomalies of the corresponding continuum theory. An important role is played by the lattice translation by one lattice site, which in the continuum limit involves a discrete axial transformation. On a lattice with periodic boundary conditions, the Hilbert space breaks up into sectors labeled by the Nc-ality p = 0, … Nc − 1. Our symmetry analysis implies various exact degeneracies in the spectrum of the lattice model. In particular, it shows that, for m = 0 and even Nc, the sectors p and p′ are degenerate if |p − p′| = Nc/2. In the Nc = 2 case, we explicitly construct the action of the Hamiltonian on a basis of gauge-invariant states, and we perform both a strong coupling expansion and exact diagonalization for lattices of up to 12 lattice sites. Upon extrapolation of these results, we find good agreement with the spectrum computed previously using discretized light-cone quantization. One of our new results is the first numerical calculation of the fermion bilinear condensate.

Analysis of phase shift between pulse oscillations of macro- and microvascular cerebral blood flow in patients with traumatic brain injury

PurposeAfter a traumatic brain injury (TBI), monitoring of both macrovascular and microvascular blood circulation can potentially yield a better understanding of pathophysiology of potential secondary brain lesions. We investigated the changes in phase shift (PS) between cardiac-induced oscillations of cerebral blood flow (CBF) measured at macro (ultrasound Doppler) and microvascular (laser Doppler) level. Further we assessed the impact of intracranial pressure (ICP) on PS in TBI patients. A secondary aim was to compare PS to TCD-derived cerebral arterial time constant (τ), a parameter that reflects the circulatory transit time.MethodsTCD blood flow velocities (FV) in the middle cerebral artery, laser Doppler blood microcirculation flux (LDF), arterial blood pressure (ABP), and ICP were monitored in 29 consecutive patients with TBI. Eight patients were excluded because of poor-quality signals. For the remaining 21 patients (median age = 23 (Q1: 20–Q3: 33); men:16,) data were retrospectively analysed. PS between the fundamental harmonics of FV and LDF signals was determined using spectral analysis. τ was estimated as a product of cerebrovascular resistance and compliance, based on the mathematical transformation of FV and ABP, ICP pulse waveforms.ResultsPS was negative (median: −26 (Q1: −38–Q3: −15) degrees) indicating that pulse LDF at a heart rate frequency lagged behind TCD pulse. With rising mean ICP, PS became more negative (R = −0.51, p < 0.019) indicating that delay of LDF pulse increases. There was a significant correlation between PS and cerebrovascular time constant (R = −0.47, p = 0.03).ConclusionsPulse divergence between FV and LDF became greater with elevated ICP, likely reflecting prolonged circulatory travel time.

MeshWGAN: Mesh-to-Mesh Wasserstein GAN With Multi-Task Gradient Penalty for 3D Facial Geometric Age Transformation.

As the metaverse develops rapidly, 3D facial age transformation is attracting increasing attention, which may bring many potential benefits to a wide variety of users, e.g., 3D aging figures creation, 3D facial data augmentation and editing. Compared with 2D methods, 3D face aging is an underexplored problem. To fill this gap, we propose a new mesh-to-mesh Wasserstein generative adversarial network (MeshWGAN) with a multi-task gradient penalty to model a continuous bi-directional 3D facial geometric aging process. To the best of our knowledge, this is the first architecture to achieve 3D facial geometric age transformation via real 3D scans. As previous image-to-image translation methods cannot be directly applied to the 3D facial mesh, which is totally different from 2D images, we built a mesh encoder, decoder, and multi-task discriminator to facilitate mesh-to-mesh transformations. To mitigate the lack of 3D datasets containing children's faces, we collected scans from 765 subjects aged 5-17 in combination with existing 3D face databases, which provided a large training dataset. Experiments have shown that our architecture can predict 3D facial aging geometries with better identity preservation and age closeness compared to 3D trivial baselines. We also demonstrated the advantages of our approach via various 3D face-related graphics applications.