Searcharxiv⌕ Search

arXiv subjects

Chao Yang

Publications and source records attributed to Chao Yang.

At least 325 records · Page 18Linked to original sources

AUTO3D: Novel view synthesis through unsupervisely learned variational viewpoint and global 3D representation

This paper targets on learning-based novel view synthesis from a single or limited 2D images without the pose supervision. In the viewer-centered coordinates, we construct an end-to-end trainable conditional variational framework to disentangle the unsupervisely learned relative-pose/rotation and implicit global 3D representation (shape, texture and the origin of viewer-centered coordinates, etc.). The global appearance of the 3D object is given by several appearance-describing images taken from any number of viewpoints. Our spatial correlation module extracts a global 3D representation from the appearance-describing images in a permutation invariant manner. Our system can achieve implicitly 3D understanding without explicitly 3D reconstruction. With an unsupervisely learned viewer-centered relative-pose/rotation code, the decoder can hallucinate the novel view continuously by sampling the relative-pose in a prior distribution. In various applications, we demonstrate that our model can achieve comparable or even better results than pose/3D model-supervised learning-based novel view synthesis (NVS) methods with any number of input views.

cs.CV↗

Multi-time scale identification for multi-energy system

Multi-energy systems have been leaping forward for its various benefits, e.g., energy conservation and emission reduction. Coupling components are capable of transmitting energy from one time scale system to another time scale system, so the multi-energy system exhibits multi-time scale characteristic and broad bandwidth, thereby causing difficulties in dynamic modeling. In this work, two-time scale system identification is studied. A method is developed to solve the problem, which is uses signal pre-filtering and subtraction. The high and low frequency parts of the two-time scale system are identified separately and then combined to form the incorporated in parallel structure. The consistency of the method is proved and case studies are used to verify the effectiveness of the method.

eess.SY↗

Parallel multilevel restricted Schwarz preconditioners for implicit simulation of subsurface flows with Peng-Robinson equation of state

Parallel algorithms and simulators with good scalabilities are particularly important for large-scale reservoir simulations on modern supercomputers with a large number of processors. In this paper, we introduce and study a family of highly scalable multilevel restricted additive Schwarz (RAS) methods for the fully implicit solution of subsurface flows with Peng-Robinson equation of state in two and three dimensions. With the use of a second-order fully implicit scheme, the proposed simulator is unconditionally stable with the relaxation of the time step size by the stability condition. The investigation then focuses on the development of several types of multilevel overlapping additive Schwarz methods for the preconditioning of the resultant linear system arising from the inexact Newton iteration, and some fast solver technologies are presented for the assurance of the multilevel approach efficiency and scalability. We numerically show that the proposed fully implicit framework is highly efficient for solving both standard benchmarks as well as realistic problems with several hundreds of millions of unknowns and scalable to 8192 processors on the Tianhe-2 supercomputer.

math.NA↗

Parallel energy-stable solver for a coupled Allen-Cahn and Cahn-Hilliard system

In this paper, we study numerical methods for solving the coupled Allen-Cahn/Cahn-Hilliard system associated with a free energy functional of logarithmic type. To tackle the challenge posed by the special free energy functional, we propose a method to approximate the discrete variational derivatives in polynomial forms, such that the corresponding finite difference scheme is unconditionally energy stable and the energy dissipation law is maintained. To further improve the performance of the algorithm, a modified adaptive time stepping strategy is adopted such that the time step size can be flexibly controlled based on the dynamical evolution of the problem. To achieve high performance on parallel computers, we introduce a domain decomposition based, parallel Newton-Krylov-Schwarz method to solve the nonlinear algebraic system constructed from the discretization at each time step. Numerical experiments show that the proposed algorithm is second-order accurate in both space and time, energy stable with large time steps, and highly scalable to over ten thousands processor cores on the Sunway TaihuLight supercomputer.

math.NA↗

Rethinking Image Inpainting via a Mutual Encoder-Decoder with Feature Equalizations

Deep encoder-decoder based CNNs have advanced image inpainting methods for hole filling. While existing methods recover structures and textures step-by-step in the hole regions, they typically use two encoder-decoders for separate recovery. The CNN features of each encoder are learned to capture either missing structures or textures without considering them as a whole. The insufficient utilization of these encoder features limit the performance of recovering both structures and textures. In this paper, we propose a mutual encoder-decoder CNN for joint recovery of both. We use CNN features from the deep and shallow layers of the encoder to represent structures and textures of an input image, respectively. The deep layer features are sent to a structure branch and the shallow layer features are sent to a texture branch. In each branch, we fill holes in multiple scales of the CNN features. The filled CNN features from both branches are concatenated and then equalized. During feature equalization, we reweigh channel attentions first and propose a bilateral propagation activation function to enable spatial equalization. To this end, the filled CNN features of structure and texture mutually benefit each other to represent image content at all feature levels. We use the equalized feature to supplement decoder features for output image generation through skip connections. Experiments on the benchmark datasets show the proposed method is effective to recover structures and textures and performs favorably against state-of-the-art approaches.

cs.CV↗

Optical System Design of Bionic Compound Eye with Broad Field of View

In nature, many common insects have compound eyes composed of many small eyes arranged on a curved retina. This kind of vision systems have many advantages, such as small size, large FOV (field of view) and high sensitivity, which have attracted extensive attention and research from world-wide researchers. It has good application prospects in military strikes and mechanical vision. In this paper, a new type of miniature compound eye system with large FOV is designed, which contains a micro-lens array and a relay system. Hexagonal micro-lens array are spliced seamlessly as a curved shell in the designed compound eye system. The intermediate curved image formed by the curved array is converted to a planar image by introducing a relay system. After combination and optimization of the micro-lens array and the relay system, the MTF values at 89.3lp/mm for each FOV within 120.5° are greater than 0.3, and the corresponding RMS spot radii less than the radius of the Airy disk, which proves the good imaging quality for the compound eye. The clear aperture of a single micro lens is 250μm with FOV 6°. After tolerance analysis, the results show the image quality still holds good enough performance and meets the requirements of the additive manufacturing process.

physics.optics↗

On the Efficient Evaluation of the Exchange Correlation Potential on Graphics Processing Unit Clusters

The predominance of Kohn-Sham density functional theory (KS-DFT) for the theoretical treatment of large experimentally relevant systems in molecular chemistry and materials science relies primarily on the existence of efficient software implementations which are capable of leveraging the latest advances in modern high performance computing (HPC). With recent trends in HPC leading towards in increasing reliance on heterogeneous accelerator based architectures such as graphics processing units (GPU), existing code bases must embrace these architectural advances to maintain the high-levels of performance which have come to be expected for these methods. In this work, we purpose a three-level parallelism scheme for the distributed numerical integration of the exchange-correlation (XC) potential in the Gaussian basis set discretization of the Kohn-Sham equations on large computing clusters consisting of multiple GPUs per compute node. In addition, we purpose and demonstrate the efficacy of the use of batched kernels, including batched level-3 BLAS operations, in achieving high-levels of performance on the GPU. We demonstrate the performance and scalability of the implementation of the purposed method in the NWChemEx software package by comparing to the existing scalable CPU XC integration in NWChem.

physics.comp-ph↗

ELSI -- An Open Infrastructure for Electronic Structure Solvers

Routine applications of electronic structure theory to molecules and periodic systems need to compute the electron density from given Hamiltonian and, in case of non-orthogonal basis sets, overlap matrices. System sizes can range from few to thousands or, in some examples, millions of atoms. Different discretization schemes (basis sets) and different system geometries (finite non-periodic vs. infinite periodic boundary conditions) yield matrices with different structures. The ELectronic Structure Infrastructure (ELSI) project provides an open-source software interface to facilitate the implementation and optimal use of high-performance solver libraries covering cubic scaling eigensolvers, linear scaling density-matrix-based algorithms, and other reduced scaling methods in between. In this paper, we present recent improvements and developments inside ELSI, mainly covering (1) new solvers connected to the interface, (2) matrix layout and communication adapted for parallel calculations of periodic and/or spin-polarized systems, (3) routines for density matrix extrapolation in geometry optimization and molecular dynamics calculations, and (4) general utilities such as parallel matrix I/O and JSON output. The ELSI interface has been integrated into four electronic structure code projects (DFTB+, DGDFT, FHI-aims, SIESTA), allowing us to rigorously benchmark the performance of the solvers on an equal footing. Based on results of a systematic set of large-scale benchmarks performed with Kohn-Sham density-functional theory and density-functional tight-binding theory, we identify factors that strongly affect the efficiency of the solvers, and propose a decision layer that assists with the solver selection process. Finally, we describe a reverse communication interface encoding matrix-free iterative solver strategies that are amenable, e.g., for use with planewave basis sets.

physics.comp-ph↗

A Multilevel Spectral Indicator Method for Eigenvalues of Large Non-Hermitian Matrices

Recently a novel family of eigensolvers, called spectral indicator methods (SIMs), was proposed. Given a region on the complex plane, SIMs first compute an indicator by the spectral projection. The indicator is used to test if the region contains eigenvalue(s). Then the region containing eigenvalues(s) is subdivided and tested. The procedure is repeated until the eigenvalues are identified within a specified precision. In this paper, using Cayley transformation and Krylov subspaces, a memory efficient multilevel eigensolver is proposed. The method uses less memory compared with the early versions of SIMs and is particularly suitable to compute many eigenvalues of large sparse (non-Hermitian) matrices. Several examples are presented for demonstration.

math.NA↗

Testing Ghasemi-Nodehi-Bambi metric parameters with quasi-periodic oscillations

QPOs are seen as peak features in the X-ray power spectral density of stellar mass black holes and neutron stars, of which frequencies can be measured with high precision. These QPO frequencies are believed to be related to fundamental frequencies of test particles, which are mainly determined by the background metric. We consider the metric introduced in Ghasemi-Nodehi and Bambi (Eur Phys J C 76:290, 2016). The fundamental frequencies in this metric are sensitive to some of the metric parameters but insensitive to other parameters, which means, the differences in fundamental frequencies in this metric and those in the Kerr ones can be significant for small changes of some but not all parameters around the Kerr value. By comparing with the QPO observations of GRO J1655-40, we find that only one parameter of the Ghasemi-Nodehi-Bambi metric can be strongly constrained, but other parameters cannot. We also use nested algorithm to investigate whether better constraints on the Ghasemi-Nodehi-Bambi metric parameters can be obtained from QPO observations of multiple objects by simulations. We find that four parameters can be strongly constrained while other parameters cannot . Our results suggest that QPOs may be important tools for testing the Kerr metric.

gr-qc↗

A Shift Selection Strategy for Parallel Shift-Invert Spectrum Slicing in Symmetric Self-Consistent Eigenvalue Computation

The central importance of large scale eigenvalue problems in scientific computation necessitates the development of massively parallel algorithms for their solution. Recent advances in dense numerical linear algebra have enabled the routine treatment of eigenvalue problems with dimensions on the order of hundreds of thousands on the world's largest supercomputers. In cases where dense treatments are not feasible, Krylov subspace methods offer an attractive alternative due to the fact that they do not require storage of the problem matrices. However, demonstration of scalability of either of these classes of eigenvalue algorithms on computing architectures capable of expressing massive parallelism is non-trivial due to communication requirements and serial bottlenecks, respectively. In this work, we introduce the SISLICE method: a parallel shift-invert algorithm for the solution of the symmetric self-consistent field (SCF) eigenvalue problem. The SISLICE method drastically reduces the communication requirement of current parallel shift-invert eigenvalue algorithms through various shift selection and migration techniques based on density of states estimation and k-means clustering, respectively. This work demonstrates the robustness and parallel performance of the SISLICE method on a representative set of SCF eigenvalue problems and outlines research directions which will be explored in future work.

math.NA↗

Quench dynamics of Hopf insulators

Hopf insulators are exotic topological states of matter outside the standard ten-fold way classification based on discrete symmetries. Its topology is captured by an integer invariant that describes the linking structures of the Hamiltonian in the three-dimensional momentum space. In this paper, we investigate the quantum dynamics of Hopf insulators across a sudden quench and show that the quench dynamics is characterized by a $\mathbb{Z}_2$ invariant $ν$ which reveals a rich interplay between quantum quench and static band topology. We construct the $\mathbb{Z}_2$ topological invariant using the loop unitary operator, and prove that $ν$ relates the pre- and post-quench Hopf invariants through $ν=(\mathcal{L}-\mathcal{L}_0)\bmod 2$. The $\mathbb{Z}_2$ nature of the dynamical invariant is in sharp contrast to the $\mathbb{Z}$ invariant for the quench dynamics of Chern insulators in two dimensions. The non-trivial dynamical topology is further attributed to the emergence of $π$-defects in the phase band of the loop unitary. These $π$-defects are generally closed curves in the momentum-time space, for example, as nodal rings carrying Hopf charge.

cond-mat.quant-gas↗

Fast and Robust Registration of Aerial Images and LiDAR data Based on Structrual Features and 3D Phase Correlation

Co-Registration of aerial imagery and Light Detection and Ranging (LiDAR) data is quilt challenging because the different imaging mechanism causes significant geometric and radiometric distortions between such data. To tackle the problem, this paper proposes an automatic registration method based on structural features and three-dimension (3D) phase correlation. In the proposed method, the LiDAR point cloud data is first transformed into the intensity map, which is used as the reference image. Then, we employ the Fast operator to extract uniformly distributed interest points in the aerial image by a partition strategy and perform a local geometric correction by using the collinearity equation to eliminate scale and rotation difference between images. Subsequently, a robust structural feature descriptor is build based on dense gradient features, and the 3D phase correlation is used to detect control points (CPs) between aerial images and LiDAR data in the frequency domain, where the image matching is accelerated by the 3D Fast Fourier Transform (FFT). Finally, the obtained CPs are employed to correct the exterior orientation elements, which is used to achieve co-registration of aerial images and LiDAR data. Experiments with two datasets of aerial images and LiDAR data show that the proposed method is much faster and more robust than state of the art methods

cs.CV↗

One-Shot Domain Adaptation For Face Generation

In this paper, we propose a framework capable of generating face images that fall into the same distribution as that of a given one-shot example. We leverage a pre-trained StyleGAN model that already learned the generic face distribution. Given the one-shot target, we develop an iterative optimization scheme that rapidly adapts the weights of the model to shift the output's high-level distribution to the target's. To generate images of the same distribution, we introduce a style-mixing technique that transfers the low-level statistics from the target to faces randomly generated with the model. With that, we are able to generate an unlimited number of faces that inherit from the distribution of both generic human faces and the one-shot example. The newly generated faces can serve as augmented training data for other downstream tasks. Such setting is appealing as it requires labeling very few, or even one example, in the target domain, which is often the case of real-world face manipulations that result from a variety of unknown and unique distributions, each with extremely low prevalence. We show the effectiveness of our one-shot approach for detecting face manipulations and compare it with other few-shot domain adaptation methods qualitatively and quantitatively.

cs.CV↗

Constrained R-CNN: A general image manipulation detection model

Recently, deep learning-based models have exhibited remarkable performance for image manipulation detection. However, most of them suffer from poor universality of handcrafted or predetermined features. Meanwhile, they only focus on manipulation localization and overlook manipulation classification. To address these issues, we propose a coarse-to-fine architecture named Constrained R-CNN for complete and accurate image forensics. First, the learnable manipulation feature extractor learns a unified feature representation directly from data. Second, the attention region proposal network effectively discriminates manipulated regions for the next manipulation classification and coarse localization. Then, the skip structure fuses low-level and high-level information to refine the global manipulation features. Finally, the coarse localization information guides the model to further learn the finer local features and segment out the tampered region. Experimental results show that our model achieves state-of-the-art performance. Especially, the F1 score is increased by 28.4%, 73.2%, 13.3% on the NIST16, COVERAGE, and Columbia dataset.

cs.CV↗

Context-Integrated and Feature-Refined Network for Lightweight Object Parsing

Semantic segmentation for lightweight object parsing is a very challenging task, because both accuracy and efficiency (e.g., execution speed, memory footprint or computational complexity) should all be taken into account. However, most previous works pay too much attention to one-sided perspective, either accuracy or speed, and ignore others, which poses a great limitation to actual demands of intelligent devices. To tackle this dilemma, we propose a novel lightweight architecture named Context-Integrated and Feature-Refined Network (CIFReNet). The core components of CIFReNet are the Long-skip Refinement Module (LRM) and the Multi-scale Context Integration Module (MCIM). The LRM is designed to ease the propagation of spatial information between low-level and high-level stages. Furthermore, channel attention mechanism is introduced into the process of long-skip learning to boost the quality of low-level feature refinement. Meanwhile, the MCIM consists of three cascaded Dense Semantic Pyramid (DSP) blocks with image-level features, which is presented to encode multiple context information and enlarge the field of view. Specifically, the proposed DSP block exploits a dense feature sampling strategy to enhance the information representations without significantly increasing the computation cost. Comprehensive experiments are conducted on three benchmark datasets for object parsing including Cityscapes, CamVid, and Helen. As indicated, the proposed method reaches a better trade-off between accuracy and efficiency compared with the other state-of-the-art methods.

cs.CV↗

One-Stage Inpainting with Bilateral Attention and Pyramid Filling Block

Recent deep learning based image inpainting methods which utilize contextual information and two-stage architecture have exhibited remarkable performance. However, the two-stage architecture is time-consuming, the contextual information lack high-level semantics and ignores both the semantic relevance and distance information of hole's feature patches, these limitations result in blurry textures and distorted structures of final result. Motivated by these observations, we propose a new deep generative model-based approach, which trains a shared network twice with different targets and utilizes a single network during the testing phase, so that we can effectively save inference time. Specifically, the targets of two training steps are structure reconstruction and texture generation respectively. During the second training, we first propose a Pyramid Filling Block (PF-block) to utilize the high-level features that the hole regions has been filled to guide the filling process of low-level features progressively, the missing content can be filled from deep to shallow in a pyramid fashion. Then, inspired by the classical bilateral filter [30], we propose the Bilateral Attention layer (BA-layer) to optimize filled feature map, which synthesizes feature patches at each position by computing weighted sums of the surrounding feature patches, these weights are derived by considering both distance and value relationships between feature patches, thus making the visually plausible inpainting results. Finally, experiments on multiple publicly available datasets show the superior performance of our approach.

cs.CV↗

Unconstrained Facial Expression Transfer using Style-based Generator

Facial expression transfer and reenactment has been an important research problem given its applications in face editing, image manipulation, and fabricated videos generation. We present a novel method for image-based facial expression transfer, leveraging the recent style-based GAN shown to be very effective for creating realistic looking images. Given two face images, our method can create plausible results that combine the appearance of one image and the expression of the other. To achieve this, we first propose an optimization procedure based on StyleGAN to infer hierarchical style vector from an image that disentangle different attributes of the face. We further introduce a linear combination scheme that fuses the style vectors of the two given images and generate a new face that combines the expression and appearance of the inputs. Our method can create high-quality synthesis with accurate facial reenactment. Unlike many existing methods, we do not rely on geometry annotations, and can be applied to unconstrained facial images of any identities without the need for retraining, making it feasible to generate large-scale expression-transferred results.

cs.CV↗