arXiv subjects

Zheng Fang

Publications and source records attributed to Zheng Fang.

At least 127 records · Page 7Linked to original sources

arXiv2021

3D Object Tracking with Transformer

Feature fusion and similarity computation are two core problems in 3D object tracking, especially for object tracking using sparse and disordered point clouds. Feature fusion could make similarity computing more efficient by including target object information. However, most existing LiDAR-based approaches directly use the extracted point cloud feature to compute similarity while ignoring the attention changes of object regions during tracking. In this paper, we propose a feature fusion network based on transformer architecture. Benefiting from the self-attention mechanism, the transformer encoder captures the inter- and intra- relations among different regions of the point cloud. By using cross-attention, the transformer decoder fuses features and includes more target cues into the current point cloud feature to compute the region attentions, which makes the similarity computing more efficient. Based on this feature fusion network, we propose an end-to-end point cloud object tracking framework, a simple yet effective method for 3D object tracking using point clouds. Comprehensive experimental results on the KITTI dataset show that our method achieves new state-of-the-art performance. Code is available at: https://github.com/3bobo/lttr.

cs.CV↗

arXiv2021

Unrestricted Adversarial Attacks on ImageNet Competition

Many works have investigated the adversarial attacks or defenses under the settings where a bounded and imperceptible perturbation can be added to the input. However in the real-world, the attacker does not need to comply with this restriction. In fact, more threats to the deep model come from unrestricted adversarial examples, that is, the attacker makes large and visible modifications on the image, which causes the model classifying mistakenly, but does not affect the normal observation in human perspective. Unrestricted adversarial attack is a popular and practical direction but has not been studied thoroughly. We organize this competition with the purpose of exploring more effective unrestricted adversarial attack algorithm, so as to accelerate the academical research on the model robustness under stronger unbounded attacks. The competition is held on the TianChi platform (\url{https://tianchi.aliyun.com/competition/entrance/531853/introduction}) as one of the series of AI Security Challengers Program.

cs.CV↗

arXiv2021

PTT: Point-Track-Transformer Module for 3D Single Object Tracking in Point Clouds

3D single object tracking is a key issue for robotics. In this paper, we propose a transformer module called Point-Track-Transformer (PTT) for point cloud-based 3D single object tracking. PTT module contains three blocks for feature embedding, position encoding, and self-attention feature computation. Feature embedding aims to place features closer in the embedding space if they have similar semantic information. Position encoding is used to encode coordinates of point clouds into high dimension distinguishable features. Self-attention generates refined attention features by computing attention weights. Besides, we embed the PTT module into the open-source state-of-the-art method P2B to construct PTT-Net. Experiments on the KITTI dataset reveal that our PTT-Net surpasses the state-of-the-art by a noticeable margin (~10%). Additionally, PTT-Net could achieve real-time performance (~40FPS) on NVIDIA 1080Ti GPU. Our code is open-sourced for the robotics community at https://github.com/shanjiayao/PTT.

cs.CV↗

arXiv2021

A Projection Framework for Testing Shape Restrictions That Form Convex Cones

This paper develops a uniformly valid and asymptotically nonconservative test based on projection for a class of shape restrictions. The key insight we exploit is that these restrictions form convex cones, a simple and yet elegant structure that has been barely harnessed in the literature. Based on a monotonicity property afforded by such a geometric structure, we construct a bootstrap procedure that, unlike many studies in nonstandard settings, dispenses with estimation of local parameter spaces, and the critical values are obtained in a way as simple as computing the test statistic. Moreover, by appealing to strong approximations, our framework accommodates nonparametric regression models as well as distributional/density-related and structural settings. Since the test entails a tuning parameter (due to the nonstandard nature of the problem), we propose a data-driven choice and prove its validity. Monte Carlo simulations confirm that our test works well.

econ.EM↗

arXiv2021

Inference for Large-Scale Linear Systems with Known Coefficients

This paper considers the problem of testing whether there exists a non-negative solution to a possibly under-determined system of linear equations with known coefficients. This hypothesis testing problem arises naturally in a number of settings, including random coefficient, treatment effect, and discrete choice models, as well as a class of linear programming problems. As a first contribution, we obtain a novel geometric characterization of the null hypothesis in terms of identified parameters satisfying an infinite set of inequality restrictions. Using this characterization, we devise a test that requires solving only linear programs for its implementation, and thus remains computationally feasible in the high-dimensional applications that motivate our analysis. The asymptotic size of the proposed test is shown to equal at most the nominal level uniformly over a large class of distributions that permits the number of linear equations to grow with the sample size.

econ.EM↗

arXiv2021

GeodesicEmbedding (GE): A High-Dimensional Embedding Approach for Fast Geodesic Distance Queries

In this paper, we develop a novel method for fast geodesic distance queries. The key idea is to embed the mesh into a high-dimensional space, such that the Euclidean distance in the high-dimensional space can induce the geodesic distance in the original manifold surface. However, directly solving the high-dimensional embedding problem is not feasible due to the large number of variables and the fact that the embedding problem is highly nonlinear. We overcome the challenges with two novel ideas. First, instead of taking all vertices as variables, we embed only the saddle vertices, which greatly reduces the problem complexity. We then compute a local embedding for each non-saddle vertex. Second, to reduce the large approximation error resulting from the purely Euclidean embedding, we propose a cascaded optimization approach that repeatedly introduces additional embedding coordinates with a non-Euclidean function to reduce the approximation residual. Using the precomputation data, our approach can determine the geodesic distance between any two vertices in near-constant time. Computational testing results show that our method is more desirable than previous geodesic distance queries methods.

cs.GR↗

arXiv2021

3D-SiamRPN: An End-to-End Learning Method for Real-Time 3D Single Object Tracking Using Raw Point Cloud

3D single object tracking is a key issue for autonomous following robot, where the robot should robustly track and accurately localize the target for efficient following. In this paper, we propose a 3D tracking method called 3D-SiamRPN Network to track a single target object by using raw 3D point cloud data. The proposed network consists of two subnetworks. The first subnetwork is feature embedding subnetwork which is used for point cloud feature extraction and fusion. In this subnetwork, we first use PointNet++ to extract features of point cloud from template and search branches. Then, to fuse the information of features in the two branches and obtain their similarity, we propose two cross correlation modules, named Pointcloud-wise and Point-wise respectively. The second subnetwork is region proposal network(RPN), which is used to get the final 3D bounding box of the target object based on the fusion feature from cross correlation modules. In this subnetwork, we utilize the regression and classification branches of a region proposal subnetwork to obtain proposals and scores, thus get the final 3D bounding box of the target object. Experimental results on KITTI dataset show that our method has a competitive performance in both Success and Precision compared to the state-of-the-art methods, and could run in real-time at 20.8 FPS. Additionally, experimental results on H3D dataset demonstrate that our method also has good generalization ability and could achieve good tracking performance in a new scene without re-training.

cs.CV↗

arXiv2021

A Unifying Framework for Testing Shape Restrictions

This paper makes the following original contributions. First, we develop a unifying framework for testing shape restrictions based on the Wald principle. The test has asymptotic uniform size control and is uniformly consistent. Second, we examine the applicability and usefulness of some prominent shape enforcing operators in implementing our framework. In particular, in stark contrast to its use in point and interval estimation, the rearrangement operator is inapplicable due to a lack of convexity. The greatest convex minorization and the least concave majorization are shown to enjoy the analytic properties required to employ our framework. Third, we show that, despite that the projection operator may not be well-defined/behaved in general parameter spaces such as those defined by uniform norms, one may nonetheless employ a powerful distance-based test by applying our framework. Monte Carlo simulations confirm that our test works well. We further showcase the empirical relevance by investigating the relationship between weekly working hours and the annual wage growth in the high-end labor market.

econ.EM↗

arXiv2021

Implementing an Improved Test of Matrix Rank in Stata

We develop a Stata command, bootranktest, for implementing the matrix rank test of Chen and Fang (2019) in linear instrumental variable regression models. Existing rank tests employ critical values that may be too small, and hence may not even be first order valid in the sense that they may fail to control the Type I error. By appealing to the bootstrap, they devise a test that overcomes the deficiency of existing tests. The command bootranktest implements the two-step version of their test, and also the analytic version if chosen. The command also accommodates data with temporal and cluster dependence.

econ.EM↗

arXiv2021

Spatial-Temporal Graph ODE Networks for Traffic Flow Forecasting

Spatial-temporal forecasting has attracted tremendous attention in a wide range of applications, and traffic flow prediction is a canonical and typical example. The complex and long-range spatial-temporal correlations of traffic flow bring it to a most intractable challenge. Existing works typically utilize shallow graph convolution networks (GNNs) and temporal extracting modules to model spatial and temporal dependencies respectively. However, the representation ability of such models is limited due to: (1) shallow GNNs are incapable to capture long-range spatial correlations, (2) only spatial connections are considered and a mass of semantic connections are ignored, which are of great importance for a comprehensive understanding of traffic networks. To this end, we propose Spatial-Temporal Graph Ordinary Differential Equation Networks (STGODE). Specifically, we capture spatial-temporal dynamics through a tensor-based ordinary differential equation (ODE), as a result, deeper networks can be constructed and spatial-temporal features are utilized synchronously. To understand the network more comprehensively, semantical adjacency matrix is considered in our model, and a well-design temporal dialated convolution structure is used to capture long term temporal dependencies. We evaluate our model on multiple real-world traffic datasets and superior performance is achieved over state-of-the-art baselines.

cs.LG↗

arXiv2021

A Query-Driven Topic Model

Topic modeling is an unsupervised method for revealing the hidden semantic structure of a corpus. It has been increasingly widely adopted as a tool in the social sciences, including political science, digital humanities and sociological research in general. One desirable property of topic models is to allow users to find topics describing a specific aspect of the corpus. A possible solution is to incorporate domain-specific knowledge into topic modeling, but this requires a specification from domain experts. We propose a novel query-driven topic model that allows users to specify a simple query in words or phrases and return query-related topics, thus avoiding tedious work from domain experts. Our proposed approach is particularly attractive when the user-specified query has a low occurrence in a text corpus, making it difficult for traditional topic models built on word cooccurrence patterns to identify relevant topics. Experimental results demonstrate the effectiveness of our model in comparison with both classical topic models and neural topic models.

cs.IR↗

arXiv2021

NMS-Loss: Learning with Non-Maximum Suppression for Crowded Pedestrian Detection

Non-Maximum Suppression (NMS) is essential for object detection and affects the evaluation results by incorporating False Positives (FP) and False Negatives (FN), especially in crowd occlusion scenes. In this paper, we raise the problem of weak connection between the training targets and the evaluation metrics caused by NMS and propose a novel NMS-Loss making the NMS procedure can be trained end-to-end without any additional network parameters. Our NMS-Loss punishes two cases when FP is not suppressed and FN is wrongly eliminated by NMS. Specifically, we propose a pull loss to pull predictions with the same target close to each other, and a push loss to push predictions with different targets away from each other. Experimental results show that with the help of NMS-Loss, our detector, namely NMS-Ped, achieves impressive results with Miss Rate of 5.92% on Caltech dataset and 10.08% on CityPersons dataset, which are both better than state-of-the-art competitors.

cs.CV↗

arXiv2021

Adiabatic Duality: Duality of cosmological models with varying slow-roll parameter and sound speed

There have been thousands of cosmological models for our early universe proposed in the literature, and many of them claimed to be able to give rise to scale-invariant power spectrum as was favored by the observational data. It is thus interesting to think about whether there are some relations among them, e.g., the duality relation. In this paper, we investigate duality relations between cosmological models in framework of general relativity (GR) , not only with varying slow-roll parameter $ε$, but also with sound speed $c_s$, which can then be understood as "adiabatic duality". Several duality relationships are formulated analytically and verified numerically. We show that models with varying $ε$ and constant $c_s$ can be dual in scalar spectral index, but not tensor one. On the other hand, allowing both $ε$ and $c_s$ to vary can make models dual in both scalar and tensor spectral indices.

gr-qc↗

arXiv2021

Improved Signed Distance Function for 2D Real-time SLAM and Accurate Localization

Accurate mapping and localization are very important for many industrial robotics applications. In this paper, we propose an improved Signed Distance Function (SDF) for both 2D SLAM and pure localization to improve the accuracy of mapping and localization. To achieve this goal, firstly we improved the back-end mapping to build a more accurate SDF map by extending the update range and building free space, etc. Secondly, to get more accurate pose estimation for the front-end, we proposed a new iterative registration method to align the current scan to the SDF submap by removing random outliers of laser scanners. Thirdly, we merged all the SDF submaps to produce an integrated SDF map for highly accurate pure localization. Experimental results show that based on the merged SDF map, a localization accuracy of a few millimeters (5mm) can be achieved globally within the map. We believe that this method is important for mobile robots working in scenarios where high localization accuracy matters.

cs.RO↗

arXiv2020

TP-TIO: A Robust Thermal-Inertial Odometry with Deep ThermalPoint

To achieve robust motion estimation in visually degraded environments, thermal odometry has been an attraction in the robotics community. However, most thermal odometry methods are purely based on classical feature extractors, which is difficult to establish robust correspondences in successive frames due to sudden photometric changes and large thermal noise. To solve this problem, we propose ThermalPoint, a lightweight feature detection network specifically tailored for producing keypoints on thermal images, providing notable anti-noise improvements compared with other state-of-the-art methods. After that, we combine ThermalPoint with a novel radiometric feature tracking method, which directly makes use of full radiometric data and establishes reliable correspondences between sequential frames. Finally, taking advantage of an optimization-based visual-inertial framework, a deep feature-based thermal-inertial odometry (TP-TIO) framework is proposed and evaluated thoroughly in various visually degraded environments. Experiments show that our method outperforms state-of-the-art visual and laser odometry methods in smoke-filled environments and achieves competitive accuracy in normal environments.

cs.RO↗

arXiv2020

TSSRGCN: Temporal Spectral Spatial Retrieval Graph Convolutional Network for Traffic Flow Forecasting

Traffic flow forecasting is of great significance for improving the efficiency of transportation systems and preventing emergencies. Due to the highly non-linearity and intricate evolutionary patterns of short-term and long-term traffic flow, existing methods often fail to take full advantage of spatial-temporal information, especially the various temporal patterns with different period shifting and the characteristics of road segments. Besides, the globality representing the absolute value of traffic status indicators and the locality representing the relative value have not been considered simultaneously. This paper proposes a neural network model that focuses on the globality and locality of traffic networks as well as the temporal patterns of traffic data. The cycle-based dilated deformable convolution block is designed to capture different time-varying trends on each node accurately. Our model can extract both global and local spatial information since we combine two graph convolutional network methods to learn the representations of nodes and edges. Experiments on two real-world datasets show that the model can scrutinize the spatial-temporal correlation of traffic data, and its performance is better than the compared state-of-the-art methods. Further analysis indicates that the locality and globality of the traffic networks are critical to traffic flow prediction and the proposed TSSRGCN model can adapt to the various temporal traffic patterns.

cs.LG↗

arXiv2020

Clinical Micro-CT Empowered by Interior Tomography, Robotic Scanning, and Deep Learning

While micro-CT systems are instrumental in preclinical research, clinical micro-CT imaging has long been desired with cochlear implantation as a primary example. The structural details of the cochlear implant and the temporal bone require a significantly higher image resolution than that (about 0.2 mm) provided by current medical CT scanners. In this paper, we propose a clinical micro-CT (CMCT) system design integrating conventional spiral cone-beam CT, contemporary interior tomography, deep learning techniques, and technologies of micro-focus X-ray source, photon-counting detector (PCD), and robotic arms for ultrahigh resolution localized tomography of a freely-selected volume of interest (VOI) at a minimized radiation dose level. The whole system consists of a standard CT scanner for a clinical CT exam and VOI specification, and a robotic-arm based micro-CT scanner for a local scan at much higher spatial and spectral resolution as well as much reduced radiation dose. The prior information from global scan is also fully utilized for background compensation to improve interior tomography from local data for accurate and stable VOI reconstruction. Our results and analysis show that the proposed hybrid reconstruction algorithm delivers superior local reconstruction, being insensitive to the misalignment of the isocenter position and initial view angle in the data/image registration while the attenuation error caused by scale mismatch can be effectively addressed with bias correction. These findings demonstrate the feasibility of our system design. We envision that deep learning techniques can be leveraged for optimized imaging performance. With high resolution imaging, high dose efficiency and low system cost synergistically, our proposed CMCT system has great potentials in temporal bone imaging as well as various other clinical applications.

physics.med-ph↗

arXiv2020

Event-VPR: End-to-End Weakly Supervised Network Architecture for Event-based Visual Place Recognition

Traditional visual place recognition (VPR) methods generally use frame-based cameras, which is easy to fail due to dramatic illumination changes or fast motions. In this paper, we propose an end-to-end visual place recognition network for event cameras, which can achieve good place recognition performance in challenging environments. The key idea of the proposed algorithm is firstly to characterize the event streams with the EST voxel grid, then extract features using a convolution network, and finally aggregate features using an improved VLAD network to realize end-to-end visual place recognition using event streams. To verify the effectiveness of the proposed algorithm, we compare the proposed method with classical VPR methods on the event-based driving datasets (MVSEC, DDD17) and the synthetic datasets (Oxford RobotCar). Experimental results show that the proposed method can achieve much better performance in challenging scenarios. To our knowledge, this is the first end-to-end event-based VPR method. The accompanying source code is available at https://github.com/kongdelei/Event-VPR.

cs.CV↗