arXiv subjects

Jian Sun

Publications and source records attributed to Jian Sun.

At least 469 records · Page 26Linked to original sources

arXiv2016

Instance-sensitive Fully Convolutional Networks

Fully convolutional networks (FCNs) have been proven very successful for semantic segmentation, but the FCN outputs are unaware of object instances. In this paper, we develop FCNs that are capable of proposing instance-level segment candidates. In contrast to the previous FCN that generates one score map, our FCN is designed to compute a small set of instance-sensitive score maps, each of which is the outcome of a pixel-wise classifier of a relative position to instances. On top of these instance-sensitive score maps, a simple assembling module is able to output instance candidate at each position. In contrast to the recent DeepMask method for segmenting instances, our method does not have any high-dimensional layer related to the mask resolution, but instead exploits image local coherence for estimating instances. We present competitive results of instance segment proposal on both PASCAL VOC and MS COCO.

cs.CV↗

arXiv2016

Effective Clipart Image Vectorization Through Direct Optimization of Bezigons

Bezigons, i.e., closed paths composed of Bézier curves, have been widely employed to describe shapes in image vectorization results. However, most existing vectorization techniques infer the bezigons by simply approximating an intermediate vector representation (such as polygons). Consequently, the resultant bezigons are sometimes imperfect due to accumulated errors, fitting ambiguities, and a lack of curve priors, especially for low-resolution images. In this paper, we describe a novel method for vectorizing clipart images. In contrast to previous methods, we directly optimize the bezigons rather than using other intermediate representations; therefore, the resultant bezigons are not only of higher fidelity compared with the original raster image but also more reasonable because they were traced by a proficient expert. To enable such optimization, we have overcome several challenges and have devised a differentiable data energy as well as several curve-based prior terms. To improve the efficiency of the optimization, we also take advantage of the local control property of bezigons and adopt an overlapped piecewise optimization strategy. The experimental results show that our method outperforms both the current state-of-the-art method and commonly used commercial software in terms of bezigon quality.

cs.GR↗

arXiv2016

Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks

State-of-the-art object detection networks depend on region proposal algorithms to hypothesize object locations. Advances like SPPnet and Fast R-CNN have reduced the running time of these detection networks, exposing region proposal computation as a bottleneck. In this work, we introduce a Region Proposal Network (RPN) that shares full-image convolutional features with the detection network, thus enabling nearly cost-free region proposals. An RPN is a fully convolutional network that simultaneously predicts object bounds and objectness scores at each position. The RPN is trained end-to-end to generate high-quality region proposals, which are used by Fast R-CNN for detection. We further merge RPN and Fast R-CNN into a single network by sharing their convolutional features---using the recently popular terminology of neural networks with 'attention' mechanisms, the RPN component tells the unified network where to look. For the very deep VGG-16 model, our detection system has a frame rate of 5fps (including all steps) on a GPU, while achieving state-of-the-art object detection accuracy on PASCAL VOC 2007, 2012, and MS COCO datasets with only 300 proposals per image. In ILSVRC and COCO 2015 competitions, Faster R-CNN and RPN are the foundations of the 1st-place winning entries in several tracks. Code has been made publicly available.

cs.CV↗

arXiv2015

Instance-aware Semantic Segmentation via Multi-task Network Cascades

Semantic segmentation research has recently witnessed rapid progress, but many leading methods are unable to identify object instances. In this paper, we present Multi-task Network Cascades for instance-aware semantic segmentation. Our model consists of three networks, respectively differentiating instances, estimating masks, and categorizing objects. These networks form a cascaded structure, and are designed to share their convolutional features. We develop an algorithm for the nontrivial end-to-end training of this causal, cascaded structure. Our solution is a clean, single-step training framework and can be generalized to cascades that have more stages. We demonstrate state-of-the-art instance-aware semantic segmentation accuracy on PASCAL VOC. Meanwhile, our method takes only 360ms testing an image using VGG-16, which is two orders of magnitude faster than previous systems for this challenging problem. As a by product, our method also achieves compelling object detection results which surpass the competitive Fast/Faster R-CNN systems. The method described in this paper is the foundation of our submissions to the MS COCO 2015 segmentation competition, where we won the 1st place.

cs.CV↗

arXiv2015

Deep Residual Learning for Image Recognition

Deeper neural networks are more difficult to train. We present a residual learning framework to ease the training of networks that are substantially deeper than those used previously. We explicitly reformulate the layers as learning residual functions with reference to the layer inputs, instead of learning unreferenced functions. We provide comprehensive empirical evidence showing that these residual networks are easier to optimize, and can gain accuracy from considerably increased depth. On the ImageNet dataset we evaluate residual nets with a depth of up to 152 layers---8x deeper than VGG nets but still having lower complexity. An ensemble of these residual nets achieves 3.57% error on the ImageNet test set. This result won the 1st place on the ILSVRC 2015 classification task. We also present analysis on CIFAR-10 with 100 and 1000 layers. The depth of representations is of central importance for many visual recognition tasks. Solely due to our extremely deep representations, we obtain a 28% relative improvement on the COCO object detection dataset. Deep residual nets are foundations of our submissions to ILSVRC & COCO 2015 competitions, where we also won the 1st places on the tasks of ImageNet detection, ImageNet localization, COCO detection, and COCO segmentation.

cs.CV↗

arXiv2015

Accelerating Very Deep Convolutional Networks for Classification and Detection

This paper aims to accelerate the test-time computation of convolutional neural networks (CNNs), especially very deep CNNs that have substantially impacted the computer vision community. Unlike previous methods that are designed for approximating linear filters or linear responses, our method takes the nonlinear units into account. We develop an effective solution to the resulting nonlinear optimization problem without the need of stochastic gradient descent (SGD). More importantly, while previous methods mainly focus on optimizing one or two layers, our nonlinear method enables an asymmetric reconstruction that reduces the rapidly accumulated error when multiple (e.g., >=10) layers are approximated. For the widely used very deep VGG-16 model, our method achieves a whole-model speedup of 4x with merely a 0.3% increase of top-5 error in ImageNet classification. Our 4x accelerated VGG-16 model also shows a graceful accuracy degradation for object detection when plugged into the Fast R-CNN detector.

cs.CV↗

arXiv2015

A new topological semimetal with iso-energetic Weyl fermions in TaAs under high pressure

TaAs as one of the experimentally discovered topological Weyl semimetal has attracted intense interests recently. The ambient TaAs has two types of Weyl nodes which are not on the same energy level. As an effective way to tune lattice parameters and electronic interactions, high pressure is becoming a significant tool to explore new materials as well as their exotic states. Therefore, it is highly interesting to investigate the behaviors of topological Weyl fermions and possible structural phase transitions in TaAs under pressure. Here, with a combination of ab initio calculations and crystal structure prediction techniques, a new hexagonal P-6m2 phase is predicted in TaAs at pressure around 14 GPa. Surprisingly, this new phase is a topological semimetal with only single set of Weyl nodes exactly on the same energy level. The phase transition pressure from the experimental measurements, including electrical transport measurements and Raman spectroscopy, agrees with our theoretical prediction reasonably. Moreover, the P-6m2 phase seems to be quenched recoverable to ambient pressure, which increases the possibilities of further study on the exotic behaviors of single set of Weyl fermions, such as the interplay between surface states and other properties.

cond-mat.mtrl-sci↗

arXiv2015

Deep Representation of Facial Geometric and Photometric Attributes for Automatic 3D Facial Expression Recognition

In this paper, we present a novel approach to automatic 3D Facial Expression Recognition (FER) based on deep representation of facial 3D geometric and 2D photometric attributes. A 3D face is firstly represented by its geometric and photometric attributes, including the geometry map, normal maps, normalized curvature map and texture map. These maps are then fed into a pre-trained deep convolutional neural network to generate the deep representation. Then the facial expression prediction is simplyachieved by training linear SVMs over the deep representation for different maps and fusing these SVM scores. The visualizations show that the deep representation provides a complete and highly discriminative coding scheme for 3D faces. Comprehensive experiments on the BU-3DFE database demonstrate that the proposed deep representation can outperform the widely used hand-crafted descriptors (i.e., LBP, SIFT, HOG, Gabor) and the state-of-art approaches under the same experimental protocols.

cs.CV↗

arXiv2015

Localization and Orbital Selectivity in Iron-Based Superconductors with Cu Substitution

We study an inhomogeneous three-orbital Hubbard model for the Cu-substituted iron pnictides using an extended real-space Green's function method combined with density functional calculations. We find that the onsite interactions of the Cu ions are the principal determinant of whether an electron dopant or a hole dopant is caused by the Cu substitution. It is found that the Cu substitution could lead to a hole doping when its onsite interactions are smaller than a critical value, as opposed to an electron doping when the interactions of Cu ions are larger than the critical value, which may explain why the effects of Cu substitution on the carrier density are entirely different in NaFe$_{1-x}$Cu$_x$As and Ba(Fe$_{1-x}$Cu$_x$)$_2$As$_2$. We also find that the effect of a doping-induced disorder is considerable in the Cu-substituted iron pnictides, and its cooperative effect with electron correlations contributes to the orbital-selective insulating phases in NaFe$_{1-x}$Cu$_x$As and Ba(Fe$_{1-x}$Cu$_x$)$_2$As$_2$.

cond-mat.supr-con↗

arXiv2015

Harmonic Extension

In this paper, we consider the harmonic extension problem, which is widely used in many applications of machine learning. We find that the transitional method of graph Laplacian fails to produce a good approximation of the classical harmonic function. To tackle this problem, we propose a new method called the point integral method (PIM). We consider the harmonic extension problem from the point of view of solving PDEs on manifolds. The basic idea of the PIM method is to approximate the harmonicity using an integral equation, which is easy to be discretized from points. Based on the integral equation, we explain the reason why the transitional graph Laplacian may fail to approximate the harmonicity in the classical sense and propose a different approach which we call the volume constraint method (VCM). Theoretically, both the PIM and the VCM computes a harmonic function with convergence guarantees, and practically, they are both simple, which amount to solve a linear system. One important application of the harmonic extension in machine learning is semi-supervised learning. We run a popular semi-supervised learning algorithm by Zhu et al. over a couple of well-known datasets and compare the performance of the aforementioned approaches. Our experiments show the PIM performs the best.

cs.LG↗

arXiv2015

Pressure-induced semimetal to superconductor transition in a three-dimensional topological material ZrTe5

As a new type of topological materials, ZrTe5 shows many exotic properties under extreme conditions. Utilizing resistance and ac magnetic susceptibility measurements under high pressure, while the resistance anomaly near 128 K is completely suppressed at 6.2 GPa, a fully superconducting transition emerges surprisingly. The superconducting transition temperature Tc increases with applied pressure, and reaches a maximum of 4.0 K at 14.6 GPa, followed by a slight drop but remaining almost constant value up to 68.5 GPa. At pressures above 21.2 GPa, a second superconducting phase with the maximum Tc of about 6.0 K appears and coexists with the original one to the maximum pressure studied in this work. In situ high-pressure synchrotron X-ray diffraction and Raman spectroscopy combined with theoretical calculations indicate the observed two-stage superconducting behavior is correlated to the structural phase transition from ambient Cmcm phase to high-pressure C2/m phase around 6 GPa, and to a mixture of two high-pressure phases of C2/m and P-1 above 20 GPa. The combination of structure, transport measurement and theoretical calculations enable a complete understanding of the emerging exotic properties in three-dimensional topological materials happened under extreme environments.

cond-mat.supr-con↗

arXiv2015

The large unsaturated magnetoresistance of Weyl semimetals

The Weyl semimetal (WSM) is a novel topological gapless state with promises exotic transport due to chiral anomaly. Recently, a family of nonmagnetic WSM candidates including TaAs, NbAs, NbP etc is confirmed by first principle calculation and experiments. The TaAs family are reported to display the large unsaturated magnetoresistance (XMR), which have not yet been explained. Here, we give a theoretical calculation of XMR based on the extended effective-medium approach to Weyl semimetals. We predict the power law of XMR at high magnetic field and the "turn-on" magnetic field, which are well in agreement with experiments data. Furthermore, we investigate the $θ$-dependence magnetoresistance and find the transition between the postive XMR and the negative magnetoresistance induced by chiral anomaly, which should be confirmed by further experiments.

cond-mat.mes-hall↗

arXiv2015

Point Integral Method for Solving Poisson-type Equations on Manifolds from Point Clouds with Convergence Guarantees

Partial differential equations (PDE) on manifolds arise in many areas, including mathematics and many applied fields. Among all kinds of PDEs, the Poisson-type equations including the standard Poisson equation and the related eigenproblem of the Laplace-Beltrami operator are of the most important. Due to the complicated geometrical structure of the manifold, it is difficult to get efficient numerical method to solve PDE on manifold. In the paper, we propose a method called point integral method (PIM) to solve the Poisson-type equations from point clouds with convergence guarantees. In PIM, the key idea is to derive the integral equations which approximates the Poisson-type equations and contains no derivatives but only the values of the unknown function. The latter makes the integral equation easy to be approximated from point cloud. In the paper, we explain the derivation of the integral equations, describe the point integral method and its implementation, and present the numerical experiments to demonstrate the convergence of PIM.

math.NA↗

arXiv2015

Observation of Electrically Tunable van der Waals Interaction in Graphene-Molecule Complex

van der Waals (vdW) interaction plays a fundamental role in the surface-molecules related phenomena. Tuning of the correlated charge fluctuation in the vdW complex is a plausible way to modulate the molecules interaction at the atomic surface. We report vdW interaction tunability of the graphene-CO$_2$ complex by combining the first principle calculations with the vdW exchange correlation density functionals and the time evaluation measurements of CO$_2$ molecules adsorption/desorption on graphene under an external electric field. The field-dependent charge transfer within the complex unveils the controllable tuning of CO$_2$ from acceptor to donor. Meanwhile the configuration of the adsorbed molecule - the equilibrium distance from graphene and O-C-O bonding angle - is modified accordingly. The range of electrical tunability is a unique feature for each type of molecules.

cond-mat.mes-hall↗

arXiv2015

Convergence of Laplacian Spectra from Point Clouds

The spectral structure of the Laplacian-Beltrami operator (LBO) on manifolds has been widely used in many applications, include spectral clustering, dimensionality reduction, mesh smoothing, compression and editing, shape segmentation, matching and parameterization, and so on. Typically, the underlying Riemannian manifold is unknown and often given by a set of sample points. The spectral structure of the LBO is estimated from some discrete Laplace operator constructed from this set of sample points. In our previous papers, we proposed the point integral method to discretize the LBO from point clouds, which is also capable to solve the eigenproblem. Then one fundmental issue is the convergence of the eigensystem of the discrete Laplacian to that of the LBO. In this paper, for compact manifolds isometrically embedded in Euclidean spaces possibly with boundary, we show that the eigenvalues and the eigenvectors obtained by the point integral method converges to the eigenvalues and the eigenfunctions of the LBO with the Neumann boundary, and in addition, we give an estimate of the convergence rate. This result provides a solid mathematical foundation for the point integral method in the computation of Laplacian spectra from point clouds.

math.NA↗

arXiv2015

BoxSup: Exploiting Bounding Boxes to Supervise Convolutional Networks for Semantic Segmentation

Recent leading approaches to semantic segmentation rely on deep convolutional networks trained with human-annotated, pixel-level segmentation masks. Such pixel-accurate supervision demands expensive labeling effort and limits the performance of deep networks that usually benefit from more training data. In this paper, we propose a method that achieves competitive accuracy but only requires easily obtained bounding box annotations. The basic idea is to iterate between automatically generating region proposals and training convolutional networks. These two steps gradually recover segmentation masks for improving the networks, and vise versa. Our method, called BoxSup, produces competitive results supervised by boxes only, on par with strong baselines fully supervised by masks under the same setting. By leveraging a large amount of bounding boxes, BoxSup further unleashes the power of deep convolutional networks and yields state-of-the-art results on PASCAL VOC 2012 and PASCAL-CONTEXT.

cs.CV↗

arXiv2015

Fast Guided Filter

The guided filter is a technique for edge-aware image filtering. Because of its nice visual quality, fast speed, and ease of implementation, the guided filter has witnessed various applications in real products, such as image editing apps in phones and stereo reconstruction, and has been included in official MATLAB and OpenCV. In this note, we remind that the guided filter can be simply sped up from O(N) time to O(N/s^2) time for a subsampling ratio s. In a variety of applications, this leads to a speedup of >10x with almost no visible degradation. We hope this acceleration will improve performance of current applications and further popularize this filter. Code is released.

cs.CV↗

arXiv2015

Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition

Existing deep convolutional neural networks (CNNs) require a fixed-size (e.g., 224x224) input image. This requirement is "artificial" and may reduce the recognition accuracy for the images or sub-images of an arbitrary size/scale. In this work, we equip the networks with another pooling strategy, "spatial pyramid pooling", to eliminate the above requirement. The new network structure, called SPP-net, can generate a fixed-length representation regardless of image size/scale. Pyramid pooling is also robust to object deformations. With these advantages, SPP-net should in general improve all CNN-based image classification methods. On the ImageNet 2012 dataset, we demonstrate that SPP-net boosts the accuracy of a variety of CNN architectures despite their different designs. On the Pascal VOC 2007 and Caltech101 datasets, SPP-net achieves state-of-the-art classification results using a single full-image representation and no fine-tuning. The power of SPP-net is also significant in object detection. Using SPP-net, we compute the feature maps from the entire image only once, and then pool features in arbitrary regions (sub-images) to generate fixed-length representations for training the detectors. This method avoids repeatedly computing the convolutional features. In processing test images, our method is 24-102x faster than the R-CNN method, while achieving better or comparable accuracy on Pascal VOC 2007. In ImageNet Large Scale Visual Recognition Challenge (ILSVRC) 2014, our methods rank #2 in object detection and #3 in image classification among all 38 teams. This manuscript also introduces the improvement made for this competition.

cs.CV↗