arXiv subjects

Hui Kong

Publications and source records attributed to Hui Kong.

54 records · Page 3Linked to original sources

arXiv2022

Unsupervised Visible-light Images Guided Cross-Spectrum Depth Estimation from Dual-Modality Cameras

Cross-spectrum depth estimation aims to provide a depth map in all illumination conditions with a pair of dual-spectrum images. It is valuable for autonomous vehicle applications when the vehicle is equipped with two cameras of different modalities. However, images captured by different-modality cameras can be photometrically quite different. Therefore, cross-spectrum depth estimation is a very challenging problem. Moreover, the shortage of large-scale open-source datasets also retards further research in this field. In this paper, we propose an unsupervised visible-light image guided cross-spectrum (i.e., thermal and visible-light, TIR-VIS in short) depth estimation framework given a pair of RGB and thermal images captured from a visible-light camera and a thermal one. We first adopt a base depth estimation network using RGB-image pairs. Then we propose a multi-scale feature transfer network to transfer features from the TIR-VIS domain to the VIS domain at the feature level to fit the trained depth estimation network. At last, we propose a cross-spectrum depth cycle consistency to improve the depth result of dual-spectrum image pairs. Meanwhile, we release a large dual-spectrum depth estimation dataset with visible-light and far-infrared stereo images captured in different scenes to the society. The experiment result shows that our method achieves better performance than the compared existing methods. Our datasets is available at https://github.com/whitecrow1027/VIS-TIR-Datasets.

cs.CV↗

arXiv2022

Graph Gain: A Concave-Hull Based Volumetric Gain for Robotic Exploration

The existing volumetric gain for robotic exploration is calculated in the 3D occupancy map, while the sampling-based exploration method is extended in the reachable (free) space. The inconsistency between them makes the existing calculation of volumetric gain inappropriate for a complete exploration of the environment. To address this issue, we propose a concave-hull based volumetric gain in a sampling-based exploration framework. The concave hull is constructed based on the viewpoints generated by Rapidly-exploring Random Tree (RRT) and the nodes that fail to expand. All space outside this concave hull is considered unknown. The volumetric gain is calculated based on the viewpoints configuration rather than using the occupancy map. With the new volumetric gain, robots can avoid inefficient or even erroneous exploration behavior caused by the inappropriateness of existing volumetric gain calculation methods. Our exploration method is evaluated against the existing state-of-the-art RRT-based method in a benchmark environment. In the evaluated environment, the average running time of our method is about 38.4% of the existing state-of-the-art method and our method is more robust.

cs.RO↗

arXiv2022

Ada-Detector: Adaptive Frontier Detector for Rapid Exploration

In this paper, we propose an efficient frontier detector method based on adaptive Rapidly-exploring Random Tree (RRT) for autonomous robot exploration. Robots can achieve real-time incremental frontier detection when they are exploring unknown environments. First, our detector adaptively adjusts the sampling space of RRT by sensing the surrounding environment structure. The adaptive sampling space can greatly improve the successful sampling rate of RRT (the ratio of the number of samples successfully added to the RRT tree to the number of sampling attempts) according to the environment structure and control the expansion bias of the RRT. Second, by generating non-uniform distributed samples, our method also solves the over-sampling problem of RRT in the sliding windows, where uniform random sampling causes over-sampling in the overlap area between two adjacent sliding windows. In this way, our detector is more inclined to sample in the latest explored area, which improves the efficiency of frontier detection and achieves incremental detection. We validated our method in three simulated benchmark scenarios. The experimental comparison shows that we reduce the frontier detection runtime by about 40% compared with the SOTA method, DSV Planner.

cs.RO↗

arXiv2022

HiTPR: Hierarchical Transformer for Place Recognition in Point Cloud

Place recognition or loop closure detection is one of the core components in a full SLAM system. In this paper, aiming at strengthening the relevancy of local neighboring points and the contextual dependency among global points simultaneously, we investigate the exploitation of transformer-based network for feature extraction, and propose a Hierarchical Transformer for Place Recognition (HiTPR). The HiTPR consists of four major parts: point cell generation, short-range transformer (SRT), long-range transformer (LRT) and global descriptor aggregation. Specifically, the point cloud is initially divided into a sequence of small cells by downsampling and nearest neighbors searching. In the SRT, we extract the local feature for each point cell. While in the LRT, we build the global dependency among all of the point cells in the whole point cloud. Experiments on several standard benchmarks demonstrate the superiority of the HiTPR in terms of average recall rate, achieving 93.71% at top 1% and 86.63% at top 1 on the Oxford RobotCar dataset for example.

cs.CV↗

arXiv2022

Learning Moving-Object Tracking with FMCW LiDAR

In this paper, we propose a learning-based moving-object tracking method utilizing our newly developed LiDAR sensor, Frequency Modulated Continuous Wave (FMCW) LiDAR. Compared with most existing commercial LiDAR sensors, our FMCW LiDAR can provide additional Doppler velocity information to each 3D point of the point clouds. Benefiting from this, we can generate instance labels as ground truth in a semi-automatic manner. Given the labels, we propose a contrastive learning framework, which pulls together the features from the same instance in embedding space and pushes apart the features from different instances, to improve the tracking quality. Extensive experiments are conducted on our recorded driving data, and the results show that our method outperforms the baseline methods by a large margin.

cs.CV↗

arXiv2021

Yaw-Guided Imitation Learning for Autonomous Driving in Urban Environments

Existing imitation learning methods suffer from low efficiency and generalization ability when facing the road option problem in an urban environment. In this paper, we propose a yaw-guided imitation learning method to improve the road option performance in an end-to-end autonomous driving paradigm in terms of the efficiency of exploiting training samples and adaptability to changing environments. Specifically, the yaw information is provided by the trajectory of the navigation map. Our end-to-end architecture, Yaw-guided Imitation Learning with ResNet34 Attention (YILRatt), integrates the ResNet34 backbone and attention mechanism to obtain an accurate perception. It does not need high precision maps and realizes fully end-to-end autonomous driving given the yaw information provided by a consumer-level GPS receiver. By analyzing the attention heat maps, we can reveal some causal relationship between decision-making and scene perception, where, in particular, failure cases are caused by erroneous perception. We collect expert experience in the Carla 0.9.11 simulator and improve the benchmark CoRL2017 and NoCrash. Experimental results show that YILRatt has a 26.27% higher success rate than the SOTA CILRS. The code, dataset, benchmark and experimental results can be found at https://github.com/Yandong024/Yaw-guided-IL.git

cs.RO↗

arXiv2021

The Completed SDSS-IV extended Baryon Oscillation Spectroscopic Survey: measurement of the BAO and growth rate of structure of the emission line galaxy sample from the anisotropic power spectrum between redshift 0.6 and 1.1

We analyse the large-scale clustering in Fourier space of emission line galaxies (ELG) from the Data Release 16 of the Sloan Digital Sky Survey IV extended Baryon Oscillation Spectroscopic Survey. The ELG sample contains 173,736 galaxies covering 1,170 square degrees in the redshift range $0.6 < z < 1.1$. We perform a BAO measurement from the post-reconstruction power spectrum monopole, and study redshift space distortions (RSD) in the first three even multipoles. Photometric variations yield fluctuations of both the angular and radial survey selection functions. Those are directly inferred from data, imposing integral constraints which we model consistently. The full data set has only a weak preference for a BAO feature ($1.4σ$). At the effective redshift $z_{\rm eff} = 0.845$ we measure $D_{\rm V}(z_{\rm eff})/r_{\rm drag} = 18.33_{-0.62}^{+0.57}$, with $D_{\rm V}$ the volume-averaged distance and $r_{\rm drag}$ the comoving sound horizon at the drag epoch. In combination with the RSD measurement, at $z_{\rm eff} = 0.85$ we find $fσ_8(z_{\rm eff}) = 0.289_{-0.096}^{+0.085}$, with $f$ the growth rate of structure and $σ_8$ the normalisation of the linear power spectrum, $D_{\rm H}(z_{\rm eff})/r_{\rm drag} = 20.0_{-2.2}^{+2.4}$ and $D_{\rm M}(z_{\rm eff})/r_{\rm drag} = 19.17 \pm 0.99$ with $D_{\rm H}$ and $D_{\rm M}$ the Hubble and comoving angular distances, respectively. These results are in agreement with those obtained in configuration space, thus allowing a consensus measurement of $fσ_8(z_{\rm eff}) = 0.315 \pm 0.095$, $D_{\rm H}(z_{\rm eff})/r_{\rm drag} = 19.6_{-2.1}^{+2.2}$ and $D_{\rm M}(z_{\rm eff})/r_{\rm drag} = 19.5 \pm 1.0$. This measurement is consistent with a flat $Λ$CDM model with Planck parameters.

astro-ph.CO↗

arXiv2021

Globally Optimal Relative Pose Estimation with Gravity Prior

Smartphones, tablets and camera systems used, e.g., in cars and UAVs, are typically equipped with IMUs (inertial measurement units) that can measure the gravity vector accurately. Using this additional information, the $y$-axes of the cameras can be aligned, reducing their relative orientation to a single degree-of-freedom. With this assumption, we propose a novel globally optimal solver, minimizing the algebraic error in the least-squares sense, to estimate the relative pose in the over-determined case. Based on the epipolar constraint, we convert the optimization problem into solving two polynomials with only two unknowns. Also, a fast solver is proposed using the first-order approximation of the rotation. The proposed solvers are compared with the state-of-the-art ones on four real-world datasets with approx. 50000 image pairs in total. Moreover, we collected a dataset, by a smartphone, consisting of 10933 image pairs, gravity directions, and ground truth 3D reconstructions.

cs.CV↗

arXiv2021

Performance of Kitt Peak's Mayall 4-meter Telescope During DESI Commissioning

In preparation for the Dark Energy Spectroscopic Instrument (DESI), a new top end was installed on the Mayall 4-meter telescope at Kitt Peak National Observatory. The refurbished telescope and the DESI instrument were successfully commissioned on sky between 2019 October and 2020 March. Here we describe the pointing, tracking and imaging performance of the Mayall telescope equipped with its new DESI prime focus corrector, as measured by six guider cameras sampling the outer edge of DESI's focal plane. Analyzing ~500,000 guider images acquired during commissioning, we find a median delivered image FWHM of 1.1 arcseconds (in the r-band at 650 nm), with the distribution extending to a best-case value of ~0.6 arcseconds. The point spread function is well characterized by a Moffat profile with a power-law index of $β$ ~ 3.5 and little dependence of $β$ on FWHM. The shape and size of the PSF delivered by the new corrector at a field angle of 1.57 degrees are very similar to those measured with the old Mayall corrector on axis. We also find that the Mayall achieves excellent pointing accuracy (several arcseconds RMS) and minimal open-loop tracking drift (< 1 milliarcsecond per second), improvements on the telecope's pre-DESI performance. In the future, employing DESI's active focus adjustment capabilities will likely further improve the Mayall/DESI delivered image quality.

astro-ph.IM↗

arXiv2020

The Completed SDSS-IV extended Baryon Oscillation Spectroscopic Survey: Large-scale Structure Catalogs for Cosmological Analysis

We present large-scale structure catalogs from the completed extended Baryon Oscillation Spectroscopic Survey (eBOSS). Derived from Sloan Digital Sky Survey (SDSS) -IV Data Release 16 (DR16), these catalogs provide the data samples, corrected for observational systematics, and random positions sampling the survey selection function. Combined, they allow large-scale clustering measurements suitable for testing cosmological models. We describe the methods used to create these catalogs for the eBOSS DR16 Luminous Red Galaxy (LRG) and Quasar samples. The quasar catalog contains 343,708 redshifts with $0.8 < z < 2.2$ over 4,808\,deg$^2$. We combine 174,816 eBOSS LRG redshifts over 4,242\,deg$^2$ in the redshift interval $0.6 < z < 1.0$ with SDSS-III BOSS LRGs in the same redshift range to produce a combined sample of 377,458 galaxy redshifts distributed over 9,493\,deg$^2$. Improved algorithms for estimating redshifts allow that 98 per cent of LRG observations result in a successful redshift, with less than one per cent catastrophic failures ($Δz > 1000$ ${\rm km~s}^{-1}$). For quasars, these rates are 95 and 2 per cent (with $Δz > 3000$ ${\rm km~s}^{-1}$). We apply corrections for trends between the number densities of our samples and the properties of the imaging and spectroscopic data. For example, the quasar catalog obtains a $χ^2$/DoF$= 776/10$ for a null test against imaging depth before corrections and a $χ^2$/DoF$=6/8$ after. The catalogs, combined with careful consideration of the details of their construction found here-in, allow companion papers to present cosmological results with negligible impact from observational systematic uncertainties.

astro-ph.CO↗

arXiv2020

Frontier Detection and Reachability Analysis for Efficient 2D Graph-SLAM Based Active Exploration

We propose an integrated approach to active exploration by exploiting the Cartographer method as the base SLAM module for submap creation and performing efficient frontier detection in the geometrically co-aligned submaps induced by graph optimization. We also carry out analysis on the reachability of frontiers and their clusters to ensure that the detected frontier can be reached by robot. Our method is tested on a mobile robot in real indoor scene to demonstrate the effectiveness and efficiency of our approach.

cs.CV↗

arXiv2020

Cascaded Non-local Neural Network for Point Cloud Semantic Segmentation

In this paper, we propose a cascaded non-local neural network for point cloud segmentation. The proposed network aims to build the long-range dependencies of point clouds for the accurate segmentation. Specifically, we develop a novel cascaded non-local module, which consists of the neighborhood-level, superpoint-level and global-level non-local blocks. First, in the neighborhood-level block, we extract the local features of the centroid points of point clouds by assigning different weights to the neighboring points. The extracted local features of the centroid points are then used to encode the superpoint-level block with the non-local operation. Finally, the global-level block aggregates the non-local features of the superpoints for semantic segmentation in an encoder-decoder framework. Benefiting from the cascaded structure, geometric structure information of different neighborhoods with the same label can be propagated. In addition, the cascaded structure can largely reduce the computational cost of the original non-local operation on point clouds. Experiments on different indoor and outdoor datasets show that our method achieves state-of-the-art performance and effectively reduces the time consumption and memory occupation.

cs.CV↗

arXiv2020

Removing Imaging Systematics from Galaxy Clustering Measurements with Obiwan : Application to the SDSS-IV extended Baryon Oscillation Spectroscopic Survey Emission Line Galaxy Sample

This work presents the application of a new tool, Obiwan , which uses image simulations to determine the selection function of a galaxy redshift survey and calculate 3-dimensional (3D) clustering statistics. This is a forward model of the process by which images of the night sky are transformed into a 3D large--scale structure catalog. The photometric pipeline automatically detects and models galaxies and then generates a catalog of such galaxies with detailed information for each one of them, including their location, redshift and so on. Systematic biases in the imaging data are therefore imparted into the catalogs and must be accounted for in any scientific analysis of their information content. Obiwan simulates this process for samples selected from the Legacy Surveys imaging data. This imaging data will be used to select target samples for the next-generation Dark Energy Spectroscopic Instrument (DESI) experiment. Here, we apply Obiwan to a portion of the SDSS-IV extend Baryon Oscillation Spectroscopic Survey Emission Line Galaxies (ELG) sample. Systematic biases in the data are clearly identified and removed. We compare the 3D clustering results to those obtained by the map--based approach applied to the full eBOSS sample. We find the results are consistent, thereby validating the eBOSS ELG catalogs, presented in Raichoor(2020), used to obtain cosmological results.

astro-ph.CO↗

arXiv2020

The completed SDSS-IV extended Baryon Oscillation Spectroscopic Survey: Large-scale Structure Catalogues and Measurement of the isotropic BAO between redshift 0.6 and 1.1 for the Emission Line Galaxy Sample

We present the Emission Line Galaxy (ELG) sample of the extended Baryon Oscillation Spectroscopic Survey (eBOSS) from the Sloan Digital Sky Survey IV Data Release 16 (DR16). After describing the observations and redshift measurement for the 269,243 observed ELG spectra over 1170 deg$^2$, we present the large-scale structure catalogues, which are used for the cosmological analysis. These catalogues contain 173,736 reliable spectroscopic redshifts between 0.6 and 1.1, along with the associated random catalogues quantifying the extent of observations, and the appropriate weights to correct for non-cosmological fluctuations. We perform a spherically averaged baryon acoustic oscillations (BAO) measurement in configuration space, with density field reconstruction: the data 2-point correlation function shows a feature consistent with that of the BAO, providing a 3.2-percent measurement of the spherically averaged BAO distance $D_V(z_{\rm eff})/r_{\rm drag} = 18.23\pm 0.58$ at the effective redshift $z_{\rm eff}=0.845$.

astro-ph.CO↗

arXiv2020

LiDAR Iris for Loop-Closure Detection

In this paper, a global descriptor for a LiDAR point cloud, called LiDAR Iris, is proposed for fast and accurate loop-closure detection. A binary signature image can be obtained for each point cloud after several LoG-Gabor filtering and thresholding operations on the LiDAR-Iris image representation. Given two point clouds, their similarities can be calculated as the Hamming distance of two corresponding binary signature images extracted from the two point clouds, respectively. Our LiDAR-Iris method can achieve a pose-invariant loop-closure detection at a descriptor level with the Fourier transform of the LiDAR-Iris representation if assuming a 3D (x,y,yaw) pose space, although our method can generally be applied to a 6D pose space by re-aligning point clouds with an additional IMU sensor. Experimental results on five road-scene sequences demonstrate its excellent performance in loop-closure detection.

cs.RO↗

arXiv2019

Piecewise Robust Barrier Tubes for Nonlinear Hybrid Systems with Uncertainty

Piecewise Barrier Tubes (PBT) is a new technique for flowpipe overapproximation for nonlinear systems with polynomial dynamics, which leverages a combination of barrier certificates. PBT has advantages over traditional time-step based methods in dealing with those nonlinear dynamical systems in which there is a large difference in speed between trajectories, producing an overapproximation that is time independent. However, the existing approach for PBT is not efficient due to the application of interval methods for enclosure-box computation, and it can only deal with continuous dynamical systems without uncertainty. In this paper, we extend the approach with the ability to handle both continuous and hybrid dynamical systems with uncertainty that can reside in parameters and/or noise. We also improve the efficiency of the method significantly, by avoiding the use of interval-based methods for the enclosure-box computation without loosing soundness. We have developed a C++ prototype implementing the proposed approach and we evaluate it on several benchmarks. The experiments show that our approach is more efficient and precise than other methods in the literature.

eess.SY↗

arXiv2016

Invariant Clusters for Hybrid Systems

In this paper, we propose an approach to automatically compute invariant clusters for semialgebraic hybrid systems. An invariant cluster for an ordinary differential equation (ODE) is a multivariate polynomial invariant g(u,x)=0, parametric in u, which can yield an infinite number of concrete invariants by assigning different values to u so that every trajectory of the system can be overapproximated precisely by a union of concrete invariants. For semialgebraic systems, which involve ODEs with multivariate polynomial vector flow, invariant clusters can be obtained by first computing the remainder of the Lie derivative of a template multivariate polynomial w.r.t. its Groebner basis and then solving the system of polynomial equations obtained from the coefficients of the remainder. Based on invariant clusters and sum-of-squares (SOS) programming, we present a new method for the safety verification of hybrid systems. Experiments on nonlinear benchmark systems from biology and control theory show that our approach is effective and efficient.

math.OC↗

arXiv2013

Exponential-Condition-Based Barrier Certificate Generation for Safety Verification of Hybrid Systems

A barrier certificate is an inductive invariant function which can be used for the safety verification of a hybrid system. Safety verification based on barrier certificate has the benefit of avoiding explicit computation of the exact reachable set which is usually intractable for nonlinear hybrid systems. In this paper, we propose a new barrier certificate condition, called Exponential Condition, for the safety verification of semi-algebraic hybrid systems. The most important benefit of Exponential Condition is that it has a lower conservativeness than the existing convex condition and meanwhile it possesses the property of convexity. On the one hand, a less conservative barrier certificate forms a tighter over-approximation for the reachable set and hence is able to verify critical safety properties. On the other hand, the property of convexity guarantees its solvability by semidefinite programming method. Some examples are presented to illustrate the effectiveness and practicality of our method.

cs.SE↗