arXiv subjects

Qian Zhang

Publications and source records attributed to Qian Zhang.

At least 379 records · Page 21Linked to original sources

arXiv2020

On Efficient and Robust Metrics for RANSAC Hypotheses and 3D Rigid Registration

This paper focuses on developing efficient and robust evaluation metrics for RANSAC hypotheses to achieve accurate 3D rigid registration. Estimating six-degree-of-freedom (6-DoF) pose from feature correspondences remains a popular approach to 3D rigid registration, where random sample consensus (RANSAC) is a de-facto choice to this problem. However, existing metrics for RANSAC hypotheses are either time-consuming or sensitive to common nuisances, parameter variations, and different application scenarios, resulting in performance deterioration in overall registration accuracy and speed. We alleviate this problem by first analyzing the contributions of inliers and outliers, and then proposing several efficient and robust metrics with different designing motivations for RANSAC hypotheses. Comparative experiments on four standard datasets with different nuisances and application scenarios verify that the proposed metrics can significantly improve the registration performance and are more robust than several state-of-the-art competitors, making them good gifts to practical applications. This work also draws an interesting conclusion, i.e., not all inliers are equal while all outliers should be equal, which may shed new light on this research problem.

cs.CV↗

arXiv2020

Kernel Two-Dimensional Ridge Regression for Subspace Clustering

Subspace clustering methods have been widely studied recently. When the inputs are 2-dimensional (2D) data, existing subspace clustering methods usually convert them into vectors, which severely damages inherent structures and relationships from original data. In this paper, we propose a novel subspace clustering method for 2D data. It directly uses 2D data as inputs such that the learning of representations benefits from inherent structures and relationships of the data. It simultaneously seeks image projection and representation coefficients such that they mutually enhance each other and lead to powerful data representations. An efficient algorithm is developed to solve the proposed objective function with provable decreasing and convergence property. Extensive experimental results verify the effectiveness of the new method.

cs.CV↗

arXiv2020

CAE-LO: LiDAR Odometry Leveraging Fully Unsupervised Convolutional Auto-Encoder for Interest Point Detection and Feature Description

As an important technology in 3D mapping, autonomous driving, and robot navigation, LiDAR odometry is still a challenging task. Appropriate data structure and unsupervised deep learning are the keys to achieve an easy adjusted LiDAR odometry solution with high performance. Utilizing compact 2D structured spherical ring projection model and voxel model which preserves the original shape of input data, we propose a fully unsupervised Convolutional Auto-Encoder based LiDAR Odometry (CAE-LO) that detects interest points from spherical ring data using 2D CAE and extracts features from multi-resolution voxel model using 3D CAE. We make several key contributions: 1) experiments based on KITTI dataset show that our interest points can capture more local details to improve the matching success rate on unstructured scenarios and our features outperform state-of-the-art by more than 50% in matching inlier ratio; 2) besides, we also propose a keyframe selection method based on matching pairs transferring, an odometry refinement method for keyframes based on extended interest points from spherical rings, and a backward pose update method. The odometry refinement experiments verify the proposed ideas' feasibility and effectiveness.

cs.CV↗

arXiv2020

Transformer Transducer: One Model Unifying Streaming and Non-streaming Speech Recognition

In this paper we present a Transformer-Transducer model architecture and a training technique to unify streaming and non-streaming speech recognition models into one model. The model is composed of a stack of transformer layers for audio encoding with no lookahead or right context and an additional stack of transformer layers on top trained with variable right context. In inference time, the context length for the variable context layers can be changed to trade off the latency and the accuracy of the model. We also show that we can run this model in a Y-model architecture with the top layers running in parallel in low latency and high latency modes. This allows us to have streaming speech recognition results with limited latency and delayed speech recognition results with large improvements in accuracy (20% relative improvement for voice-search task). We show that with limited right context (1-2 seconds of audio) and small additional latency (50-100 milliseconds) at the end of decoding, we can achieve similar accuracy with models using unlimited audio right context. We also present optimizations for audio and label encoders to speed up the inference in streaming and non-streaming speech decoding.

cs.SD↗

arXiv2020

Curl-curl conforming elements on tetrahedra

In [24], we proposed H(curl^2)-conforming elements on both a triangle and a rectangle. This family of elements provides a brand new method to solve the quad-curl problem in 2 dimensions. In this paper, we turn our focus to 3 dimensions and construct an H(curl^2)-conforming tetrahedral finite element. The newly proposed element has been proved to have the optimal interpolation error estimate. Having tetrahedral elements, we can solve the quad-curl problem in any Lipschitz domain by conforming finite element method. We also provide several numerical examples of using our element to solve the quad-curl problem. The results of the numerical experiments show the effectiveness and correctness of our element.

math.NA↗

arXiv2020

Learned Variable-Rate Multi-Frequency Image Compression using Modulated Generalized Octave Convolution

In this proposal, we design a learned multi-frequency image compression approach that uses generalized octave convolutions to factorize the latent representations into high-frequency (HF) and low-frequency (LF) components, and the LF components have lower resolution than HF components, which can improve the rate-distortion performance, similar to wavelet transform. Moreover, compared to the original octave convolution, the proposed generalized octave convolution (GoConv) and octave transposed-convolution (GoTConv) with internal activation layers preserve more spatial structure of the information, and enable more effective filtering between the HF and LF components, which further improve the performance. In addition, we develop a variable-rate scheme using the Lagrangian parameter to modulate all the internal feature maps in the auto-encoder, which allows the scheme to achieve the large bitrate range of the JPEG AI with only three models. Experiments show that the proposed scheme achieves much better Y MS-SSIM than VVC. In terms of YUV PSNR, our scheme is very similar to HEVC.

eess.IV↗

arXiv2020

AutoPose: Searching Multi-Scale Branch Aggregation for Pose Estimation

We present AutoPose, a novel neural architecture search(NAS) framework that is capable of automatically discovering multiple parallel branches of cross-scale connections towards accurate and high-resolution 2D human pose estimation. Recently, high-performance hand-crafted convolutional networks for pose estimation show growing demands on multi-scale fusion and high-resolution representations. However, current NAS works exhibit limited flexibility on scale searching, they dominantly adopt simplified search spaces of single-branch architectures. Such simplification limits the fusion of information at different scales and fails to maintain high-resolution representations. The presentedAutoPose framework is able to search for multi-branch scales and network depth, in addition to the cell-level microstructure. Motivated by the search space, a novel bi-level optimization method is presented, where the network-level architecture is searched via reinforcement learning, and the cell-level search is conducted by the gradient-based method. Within 2.5 GPU days, AutoPose is able to find very competitive architectures on the MS COCO dataset, that are also transferable to the MPII dataset. Our code is available at https://github.com/VITA-Group/AutoPose.

cs.CV↗

arXiv2020

Intensity-only Mode Decomposition on Multimode Fibers using a Densely Connected Convolutional Network

The use of multimode fibers offers advantages in the field of communication technology in terms of transferable information density and information security. For applications using physical layer security or mode division multiplexing, the complex transmission matrix must be known. To measure the transmission matrix, the individual modes of the multimode fiber are excited sequentially at the input and a mode decomposition is performed at the output. Mode decomposition is usually performed using digital holography, which requires the provision of a reference wave and leads to high efforts. To overcome these drawbacks, a neural network is proposed, which performs mode decomposition with intensity-only camera recordings of the multimode fiber facet. Due to the high computational complexity of the problem, this approach was usually limited to a number of 6 modes. In this work, it could be shown for the first time that by using a DenseNet with 121 layers it is possible to break through the hurdle of 6 modes. The advancement is demonstrated by a mode decomposition with 10 modes experimentally. The training process is based on synthetic data. The proposed method is quantitatively compared to the conventional approach with digital holography. In addition, it is shown that the network can perform mode decomposition on a 55-mode fiber, which also supports modes unknown to the neural network. The smart detection using a DenseNet opens new ways for the application of multimode fibers in optical communication networks for physical layer security.

eess.IV↗

arXiv2020

Three families of grad-div-conforming finite elements

Several smooth finite element de Rham complexes are constructed in three-dimensional space, which yield three families of grad-div conforming finite elements. The simplest element has only 8 degrees of freedom (DOFs) for a tetrahedron and 14 DOFs for a cuboid. These elements naturally lead to conforming approximations to quad-div problems. Numerical experiments for each family validate the correctness and efficiency of the elements for solving the quad-div problem.

math.NA↗

arXiv2020

Learning Where to Focus for Efficient Video Object Detection

Transferring existing image-based detectors to the video is non-trivial since the quality of frames is always deteriorated by part occlusion, rare pose, and motion blur. Previous approaches exploit to propagate and aggregate features across video frames by using optical flow-warping. However, directly applying image-level optical flow onto the high-level features might not establish accurate spatial correspondences. Therefore, a novel module called Learnable Spatio-Temporal Sampling (LSTS) has been proposed to learn semantic-level correspondences among adjacent frame features accurately. The sampled locations are first randomly initialized, then updated iteratively to find better spatial correspondences guided by detection supervision progressively. Besides, Sparsely Recursive Feature Updating (SRFU) module and Dense Feature Aggregation (DFA) module are also introduced to model temporal relations and enhance per-frame features, respectively. Without bells and whistles, the proposed method achieves state-of-the-art performance on the ImageNet VID dataset with less computational complexity and real-time speed. Code will be made available at https://github.com/jiangzhengkai/LSTS.

cs.CV↗

arXiv2020

SiamParseNet: Joint Body Parsing and Label Propagation in Infant Movement Videos

General movement assessment (GMA) of infant movement videos (IMVs) is an effective method for the early detection of cerebral palsy (CP) in infants. Automated body parsing is a crucial step towards computer-aided GMA, in which infant body parts are segmented and tracked over time for movement analysis. However, acquiring fully annotated data for video-based body parsing is particularly expensive due to the large number of frames in IMVs. In this paper, we propose a semi-supervised body parsing model, termed SiamParseNet (SPN), to jointly learn single frame body parsing and label propagation between frames in a semi-supervised fashion. The Siamese-structured SPN consists of a shared feature encoder, followed by two separate branches: one for intra-frame body parts segmentation, and one for inter-frame label propagation. The two branches are trained jointly, taking pairs of frames from the same videos as their input. An adaptive training process is proposed that alternates training modes between using input pairs of only labeled frames and using inputs of both labeled and unlabeled frames. During testing, we employ a multi-source inference mechanism, where the final result for a test frame is either obtained via the segmentation branch or via propagation from a nearby key frame. We conduct extensive experiments on a partially-labeled IMV dataset where SPN outperforms all prior arts, demonstrating the effectiveness of our proposed method.

cs.CV↗

arXiv2020

A priori and a posteriori error estimates for the quad-curl eigenvalue problem

In this paper, we propose a new family of H(curl^2)-conforming elements for the quad-curl eigenvalue problem in 2D. The accuracy of this family is one order higher than that in [32]. We prove a priori and a posteriori error estimates. The a priori estimate of the eigenvalue with a convergence order 2(s-1) is obtained if the eigenvector u\in H^{s+1}(Ω). For the a posteriori estimate, by analyzing the associated source problem, we obtain lower and upper bounds for the eigenvector in an energy norm and an upper bound for the eigenvalues. Numerical examples are presented for validation.

math.NA↗

arXiv2020

The oxygen partial pressure in solid oxide electrolysis cells with two layer electrolytes

A number of degradation mechanisms have been observed during the long-term operation of solid oxide electrolysis cells (SOEC). Using an electrolyte charge carrier transport model, we quantify the oxygen potentials across the electrolyte and thereby provide insights into these degradation mechanisms. Our model describes the transport of charge carriers in the electrolyte when the oxygen partial pressure is extremely low by accounting for the spatial variation of the concentration of oxygen vacancies in the electrolyte. Moreover, we identify four quantities that characterize the distribution of oxygen partial pressure in the electrolyte, which are directly related to the degradation mechanisms in the electrolyte as well: the two oxygen partial pressures at the interfaces of the electrodes and the electrolyte, the oxygen partial pressure at the interface of YSZ/GDC, and the position of the abrupt change in oxygen potential near the p-n junction that develops in YSZ when one side of the cell is exposed to fuel (low oxygen potential, n-type conduction) and the other side is exposed to oxidant (high oxygen potential, p-type conduction). We give analytical estimates for all of these quantities. These analytical expressions provide guidance on the parameters that need to be controlled to suppress the degradation observed in the electrolyte. In addition, the effects of operating conditions, particularly current density and operating temperature, on degradation are discussed.

physics.app-ph↗

arXiv2020

Half-Heusler thermoelectric materials: NMR studies

We report $^{59}$Co, $^{93}$Nb, and $^{121}$Sb nuclear magnetic resonance (NMR) measurements combined with density functional theory (DFT) calculations on a series of half-Heusler semiconductors, including NbCoSn, ZrCoSb, TaFeSb and NbFeSb, to better understand their electronic properties and general composition-dependent trends. These materials are of interest as potentially high efficiency thermoelectric materials. Compared to the other materials, we find that ZrCoSb tends to have a relatively large amount of local disorder, apparently antisite defects. This contributes to a small excitation gap corresponding to an impurity band near the band edge. In NbCoSn and TaFeSb, Curie-Weiss-type behavior is revealed, which indicates a small density of interacting paramagnetic defects. Very large paramagnetic chemical shifts are observed associated with a Van Vleck mechanism due to closely spaced $d$ bands splitting between the conduction and valence bands. Meanwhile, DFT methods were generally successful in reproducing the chemical shift trend for these half-Heusler materials, and we identify an enhancement of the larger-magnitude shifts, which we connect to electron interaction effects. The general trend is connected to changes in $d$-electron hybridization across the series.

cond-mat.mtrl-sci↗

arXiv2020

Role of rotational coherence in femtosecond-pulse-driven nitrogen ion lasing

We experimentally investigated the rotationally resolved polarization characteristics of N$_2^+$ lasing at 391 and 428 nm using a pump-seed scheme. By varying the relative angle between the linear polarizations of the pump and seed, it is found that the polarizations of the P and R branches of 391-nm lasing are counter-rotated. By contrast, both branches of 428-nm lasing remain polarized along the pump. The origin of the puzzled abnormal polarization characteristics is found based on a complete physical model that simultaneously includes the transient photoionization and the subsequent coupling among the electronic, vibrational and rotational quantum states of ions.It suggests that the cascaded resonant Raman processes following ionization create negative coherence between the rotational states of $J$ and $J$+2 in the ionic ground state X$^2Σ_g^+(ν=0)$, which leads to mirror-symmetrical polarization for the P and R branches of 391-nm lasing. Both the experiment and theory indicate that the demonstrated rotational coherence plays an extremely pivotal role in clarifying the gain mechanism of N$_2^+$ lasing and opens up the route toward quantum optics under ultrafast strong fields.

physics.atom-ph↗

arXiv2020

Authenticating On-Body IoT Devices: An Adversarial Learning Approach

By adding users as a new dimension to connectivity, on-body Internet-of-Things (IoT) devices have gained considerable momentum in recent years, while raising serious privacy and safety issues. Existing approaches to authenticate these devices limit themselves to dedicated sensors or specified user motions, undermining their widespread acceptance. This paper overcomes these limitations with a general authentication solution by integrating wireless physical layer (PHY) signatures with upper-layer protocols. The key enabling techniques are constructing representative radio propagation profiles from received signals, and developing an adversarial multi-player neural network to accurately recognize underlying radio propagation patterns and facilitate on-body device authentication. Once hearing a suspicious transmission, our system triggers a PHY-based challenge-response protocol to defend in depth against active attacks. We prove that at equilibrium, our adversarial model can extract all information about propagation patterns and eliminate any irrelevant information caused by motion variances and environment changes. We build a prototype of our system using Universal Software Radio Peripheral (USRP) devices and conduct extensive experiments with various static and dynamic body motions in typical indoor and outdoor environments. The experimental results show that our system achieves an average authentication accuracy of 91.6%, with a high area under the receiver operating characteristic curve (AUROC) of 0.96 and a better generalization performance compared with the conventional non-adversarial approach.

cs.CR↗

arXiv2020

VarGNet: Variable Group Convolutional Neural Network for Efficient Embedded Computing

In this paper, we propose a novel network design mechanism for efficient embedded computing. Inspired by the limited computing patterns, we propose to fix the number of channels in a group convolution, instead of the existing practice that fixing the total group numbers. Our solution based network, named Variable Group Convolutional Network (VarGNet), can be optimized easier on hardware side, due to the more unified computing schemes among the layers. Extensive experiments on various vision tasks, including classification, detection, pixel-wise parsing and face recognition, have demonstrated the practical value of our VarGNet.

cs.CV↗

arXiv2020

Densely Connected Search Space for More Flexible Neural Architecture Search

Neural architecture search (NAS) has dramatically advanced the development of neural network design. We revisit the search space design in most previous NAS methods and find the number and widths of blocks are set manually. However, block counts and block widths determine the network scale (depth and width) and make a great influence on both the accuracy and the model cost (FLOPs/latency). In this paper, we propose to search block counts and block widths by designing a densely connected search space, i.e., DenseNAS. The new search space is represented as a dense super network, which is built upon our designed routing blocks. In the super network, routing blocks are densely connected and we search for the best path between them to derive the final architecture. We further propose a chained cost estimation algorithm to approximate the model cost during the search. Both the accuracy and model cost are optimized in DenseNAS. For experiments on the MobileNetV2-based search space, DenseNAS achieves 75.3% top-1 accuracy on ImageNet with only 361MB FLOPs and 17.9ms latency on a single TITAN-XP. The larger model searched by DenseNAS achieves 76.1% accuracy with only 479M FLOPs. DenseNAS further promotes the ImageNet classification accuracies of ResNet-18, -34 and -50-B by 1.5%, 0.5% and 0.3% with 200M, 600M and 680M FLOPs reduction respectively. The related code is available at https://github.com/JaminFong/DenseNAS.

cs.CV↗