arXiv subjects

Zhiwei Li

Publications and source records attributed to Zhiwei Li.

At least 73 records · Page 4Linked to original sources

arXiv2024

SkipcrossNets: Adaptive Skip-cross Fusion for Road Detection

Multi-modal fusion is increasingly being used for autonomous driving tasks, as different modalities provide unique information for feature extraction. However, the existing two-stream networks are only fused at a specific network layer, which requires a lot of manual attempts to set up. As the CNN goes deeper, the two modal features become more and more advanced and abstract, and the fusion occurs at the feature level with a large gap, which can easily hurt the performance. To reduce the loss of height and depth information during the process of projecting point clouds into 2D space, we utilize calibration parameters to project the point cloud into Altitude Difference Images (ADIs), which exhibit more distinct road features. In this study, we propose a novel fusion architecture called Skip-cross Networks (SkipcrossNets), which combine adaptively ADIs and camera images without being bound to a certain fusion epoch. Specifically, skip-cross fusion strategy connects each layer to each layer in a feed-forward manner, and for each layer, the feature maps of all previous layers are used as input and its own feature maps are used as input to all subsequent layers for the other modality, enhancing feature propagation and multi-modal features fusion. This strategy facilitates selection of the most similar feature layers from two modalities, enhancing feature reuse and providing complementary effects for sparse point cloud features. The advantages of skip-cross fusion strategy is demonstrated through application to the KITTI and A2D2 datasets, achieving a MaxF score of 96.85% on KITTI and an F1 score of 84.84% on A2D2. The model parameters require only 2.33 MB of memory at a speed of 68.24 FPS, which can be viable for mobile terminals and embedded devices.

cs.CV↗

arXiv2024

Continuity Preserving Online CenterLine Graph Learning

Lane topology, which is usually modeled by a centerline graph, is essential for high-level autonomous driving. For a high-quality graph, both topology connectivity and spatial continuity of centerline segments are critical. However, most of existing approaches pay more attention to connectivity while neglect the continuity. Such kind of centerline graph usually cause problem to planning of autonomous driving. To overcome this problem, we present an end-to-end network, CGNet, with three key modules: 1)Junction Aware Query Enhancement module, which provides positional prior to accurately predict junction points; 2)Bézier Space Connection module, which enforces continuity constraints on any two topologically connected segments in a Bézier space; 3) Iterative Topology Refinement module, which is a graph-based network with memory to iteratively refine the predicted topological connectivity. CGNet achieves state-of-the-art performance on both nuScenes and Argoverse2 datasets.

cs.CV↗

arXiv2024

Electronic Correlation and Pseudogap-like Behavior of High-Temperature Superconductor La3Ni2O7

High-temperature superconductivity (HTSC) remains one of the most challenging and fascinating mysteries in condensed matter physics. Recently, superconductivity with transition temperature exceeding liquid-nitrogen temperature is discovered in La3Ni2O7 at high pressure, which provides a new platform to explore the unconventional HTSC. In this work, using high-resolution angle-resolved photoemission spectroscopy and ab-initio calculation, we systematically investigate the electronic structures of La3Ni2O7 at ambient pressure. Our experiments are in nice agreement with ab-initio calculations after considering an orbital-dependent band renormalization effect. The strong electron correlation effect pushes a flat band of d_(z^2 ) orbital component below the Fermi level (EF), which is predicted to locate right at EF under high pressure. Moreover, the d_(x^2-y^2 ) band shows a pseudogap-like behavior with suppressed spectral weight and diminished quasiparticle peak near EF. Our findings provide important insights into the electronic structure of La3Ni2O7, which will shed light on the understanding of the unconventional superconductivity in nickelates.

cond-mat.supr-con↗

arXiv2024

Deriva-ML: A Continuous FAIRness Approach to Reproducible Machine Learning Models

Increasingly, artificial intelligence (AI) and machine learning (ML) are used in eScience applications [9]. While these approaches have great potential, the literature has shown that ML-based approaches frequently suffer from results that are either incorrect or unreproducible due to mismanagement or misuse of data used for training and validating the models [12, 15]. Recognition of the necessity of high-quality data for correct ML results has led to data-centric ML approaches that shift the central focus from model development to creation of high-quality data sets to train and validate the models [14, 20]. However, there are limited tools and methods available for data-centric approaches to explore and evaluate ML solutions for eScience problems which often require collaborative multidisciplinary teams working with models and data that will rapidly evolve as an investigation unfolds [1]. In this paper, we show how data management tools based on the principle that all of the data for ML should be findable, accessible, interoperable and reusable (i.e. FAIR [26]) can significantly improve the quality of data that is used for ML applications. When combined with best practices that apply these tools to the entire life cycle of an ML-based eScience investigation, we can significantly improve the ability of an eScience team to create correct and reproducible ML solutions. We propose an architecture and implementation of such tools and demonstrate through two use cases how they can be used to improve ML-based eScience investigations.

cs.LG↗

arXiv2024

Nearly critical superfluid: effective field theory and holography

We study a nearly critical superfluid system from two complementary approaches. Within the first approach, we formulate a Schwinger-Keldysh effective field theory (EFT) for the system when it is located slightly above the critical temperature. . The set of symmetries, particularly the dynamical Kubo-Martin-Schwinger (KMS) symmetry and chemical shift symmetry, strictly constrains the form of EFT action. Within the second approach, using the holographic Schwinger-Keldysh technique, we derive the effective action for a ``microscopic'' holographic superfluid, confirming the EFT construction. A systematic inclusion of non-Gaussianity is one highlight of present study.

hep-th↗

arXiv2024

An Urban Water Extraction Method Combining Deep Learning and Google Earth Engine

Urban water is important for the urban ecosystem. Accurate and efficient detection of urban water with remote sensing data is of great significance for urban management and planning. In this paper, we proposed a new method to combine Google Earth Engine (GEE) with multiscale convolutional neural network (MSCNN) to extract urban water from Landsat images, which is summarized as offline training and online prediction (OTOP). That is, the training of MSCNN was completed offline, and the process of urban water extraction was implemented on GEE with the trained parameters of MSCNN. The OTOP can give full play to the respective advantages of GEE and CNN, and make the use of deep learning method on GEE more flexible. It can process available satellite images with high performance without data download and storage, and the overall performance of urban water extraction is also higher than that of the modified normalized difference water index (MNDWI) and random forest. The mean kappa, F1-score and intersection over union (IoU) of urban water extraction with the OTOP in Changchun, Wuhan, Kunming and Guangzhou reached 0.924, 0.930 and 0.869, respectively. The results of the extended validation in the other major cities of China also show that the OTOP is robust and can be used to extract different types of urban water, which benefits from the structural design and training of the MSCNN. Therefore, the OTOP is especially suitable for the study of large-scale and long-term urban water change detection in the background of urbanization.

cs.CV↗

arXiv2024

Synthesis, disorder and Ising anisotropy in a new spin liquid candidate PrMgAl$_{11}$O$_{19}$

Here we report the successful synthesis of large single crystals of triangular frustrated PrMgAl$_{11}$O$_{19}$ using the optical floating zone technique. Single crystal X-ray diffraction measurements unveiled the presence of quenched disorder within the mirror plane, specifically $\sim$7\% of Pr ions deviating from the ideal 2\textit{d} site towards the 6\textit{h} site. Magnetic susceptibility measurements revealed an Ising anisotropy with the \textit{c}-axis being the easy axis. Despite a large spin-spin interaction that develops below $\sim$10~K and considerable site disorder, the spins do not order or freeze down to at least 50 mK. The availability of large single crystals offers a distinct opportunity to investigate the exotic magnetic state on a triangular lattice with an easy axis out of the plane.

cond-mat.str-el↗

arXiv2024

Unified End-to-End V2X Cooperative Autonomous Driving

V2X cooperation, through the integration of sensor data from both vehicles and infrastructure, is considered a pivotal approach to advancing autonomous driving technology. Current research primarily focuses on enhancing perception accuracy, often overlooking the systematic improvement of accident prediction accuracy through end-to-end learning, leading to insufficient attention to the safety issues of autonomous driving. To address this challenge, this paper introduces the UniE2EV2X framework, a V2X-integrated end-to-end autonomous driving system that consolidates key driving modules within a unified network. The framework employs a deformable attention-based data fusion strategy, effectively facilitating cooperation between vehicles and infrastructure. The main advantages include: 1) significantly enhancing agents' perception and motion prediction capabilities, thereby improving the accuracy of accident predictions; 2) ensuring high reliability in the data fusion process; 3) superior end-to-end perception compared to modular approaches. Furthermore, We implement the UniE2EV2X framework on the challenging DeepAccident, a simulation dataset designed for V2X cooperative driving.

cs.CV↗

arXiv2024

SGV3D:Towards Scenario Generalization for Vision-based Roadside 3D Object Detection

Roadside perception can greatly increase the safety of autonomous vehicles by extending their perception ability beyond the visual range and addressing blind spots. However, current state-of-the-art vision-based roadside detection methods possess high accuracy on labeled scenes but have inferior performance on new scenes. This is because roadside cameras remain stationary after installation and can only collect data from a single scene, resulting in the algorithm overfitting these roadside backgrounds and camera poses. To address this issue, in this paper, we propose an innovative Scenario Generalization Framework for Vision-based Roadside 3D Object Detection, dubbed SGV3D. Specifically, we employ a Background-suppressed Module (BSM) to mitigate background overfitting in vision-centric pipelines by attenuating background features during the 2D to bird's-eye-view projection. Furthermore, by introducing the Semi-supervised Data Generation Pipeline (SSDG) using unlabeled images from new scenes, diverse instance foregrounds with varying camera poses are generated, addressing the risk of overfitting specific camera poses. We evaluate our method on two large-scale roadside benchmarks. Our method surpasses all previous methods by a significant margin in new scenes, including +42.57% for vehicle, +5.87% for pedestrian, and +14.89% for cyclist compared to BEVHeight on the DAIR-V2X-I heterologous benchmark. On the larger-scale Rope3D heterologous benchmark, we achieve notable gains of 14.48% for car and 12.41% for large vehicle. We aspire to contribute insights on the exploration of roadside perception techniques, emphasizing their capability for scenario generalization. The code will be available at https://github.com/yanglei18/SGV3D

cs.CV↗

arXiv2024

Federated Recommendation with Additive Personalization

Building recommendation systems via federated learning (FL) is a new emerging challenge for advancing next-generation Internet service and privacy protection. Existing approaches train shared item embedding by FL while keeping the user embedding private on client side. However, item embedding identical for all clients cannot capture users' individual differences on perceiving the same item and thus leads to poor personalization. Moreover, dense item embedding in FL results in expensive communication cost and latency. To address these challenges, we propose Federated Recommendation with Additive Personalization (FedRAP), which learns a global view of items via FL and a personalized view locally on each user. FedRAP enforces sparsity of the global view to save FL's communication cost and encourages difference between the two views through regularization. We propose an effective curriculum to learn the local and global views progressively with increasing regularization weights. To produce recommendations for an user, FedRAP adds the two views together to obtain a personalized item embedding. FedRAP achieves the best performance in FL setting on multiple benchmarks. It outperforms recent federated recommendation methods and several ablation study baselines.

cs.LG↗

arXiv2023

Mössbauer spectroscopy study of the magnetostructural and spin-state transitions in the breathing pyrochlore LiFeCr$_{4}$O$_{8}$

We report on investigations of the complex magnetostructural and spin-state transitions in the breathing pyrochlore LiFeCr$_{4}$O$_{8}$ by means of magnetization, Mössbauer spectroscopy, and density functional theory (DFT) calculations. Three transitions corresponding to the ferrimagnetic transition at $T_N\sim94$ K, the spin-gap transition at $T_{SG}\sim50$ K, and the magnetostructural transition at $T_{MS}\sim19$ K were observed from the $χ$(T) curve, whereas only $T_N$ and $T_{MS}$ were evidenced for the Fe site from our Mössbauer measurements, suggesting that the spin-gap transition is absent at the Fe site. This indicates that the spin-gap transition is an effect of the breathing Cr$_4$ lattice, in agreement with our DFT calculations from which we see nearly decoupled electronic states for the FeO$_4$ and CrO$_6$ units. From the temperature dependence of the hyperfine magnetic field we also observed a spin-state transition for the Fe spins at $T_{MS}$ consistent with earlier neutron diffraction measurements. These local characteristics are believed to be important for a complete understanding of the complex magnetostructural coupling effects observed in similar systems.

cond-mat.str-el↗

arXiv2023

Dual Radar: A Multi-modal Dataset with Dual 4D Radar for Autonomous Driving

Radar has stronger adaptability in adverse scenarios for autonomous driving environmental perception compared to widely adopted cameras and LiDARs. Compared with commonly used 3D radars, the latest 4D radars have precise vertical resolution and higher point cloud density, making it a highly promising sensor for autonomous driving in complex environmental perception. However, due to the much higher noise than LiDAR, manufacturers choose different filtering strategies, resulting in an inverse ratio between noise level and point cloud density. There is still a lack of comparative analysis on which method is beneficial for deep learning-based perception algorithms in autonomous driving. One of the main reasons is that current datasets only adopt one type of 4D radar, making it difficult to compare different 4D radars in the same scene. Therefore, in this paper, we introduce a novel large-scale multi-modal dataset featuring, for the first time, two types of 4D radars captured simultaneously. This dataset enables further research into effective 4D radar perception algorithms.Our dataset consists of 151 consecutive series, most of which last 20 seconds and contain 10,007 meticulously synchronized and annotated frames. Moreover, our dataset captures a variety of challenging driving scenarios, including many road conditions, weather conditions, nighttime and daytime with different lighting intensities and periods. Our dataset annotates consecutive frames, which can be applied to 3D object detection and tracking, and also supports the study of multi-modal tasks. We experimentally validate our dataset, providing valuable results for studying different types of 4D radars. This dataset is released on https://github.com/adept-thu/Dual-Radar.

cs.CV↗

arXiv2023

Compass-like manipulation of electronic nematicity in Sr$_3$Ru$_2$O$_7$

Electronic nematicity has been found in a wide range of strongly correlated electron materials, resulting in the electronic states having a symmetry that is lower than that of the crystal that hosts them. One of the most astonishing examples is Sr$_3$Ru$_2$O$_7$, in which a small in-plane component of a magnetic field induces significant resistivity anisotropy. The direction of this anisotropy follows the direction of the in-plane field. The microscopic origin of this field-induced nematicity has been a long-standing puzzle, with recent experiments suggesting a field-induced spin density wave driving the anisotropy. Here, we report spectroscopic imaging of a field-controlled anisotropy of the electronic structure at the surface of Sr$_3$Ru$_2$O$_7$. We track the electronic structure as a function of the direction of the field, revealing a continuous change with the angle. This continuous evolution suggests a mechanism based on spin-orbit coupling resulting in compass-like control of the electronic bands. The anisotropy of the electronic structure persists to temperatures about an order of magnitude higher compared to the bulk, demonstrating novel routes to stabilize such phases over a wider temperature range.

cond-mat.str-el↗

arXiv2023

Incomplete Multi-View Weak-Label Learning with Noisy Features and Imbalanced Labels

A variety of modern applications exhibit multi-view multi-label learning, where each sample has multi-view features, and multiple labels are correlated via common views. Current methods usually fail to directly deal with the setting where only a subset of features and labels are observed for each sample, and ignore the presence of noisy views and imbalanced labels in real-world problems. In this paper, we propose a novel method to overcome the limitations. It jointly embeds incomplete views and weak labels into a low-dimensional subspace with adaptive weights, and facilitates the difference between embedding weight matrices via auto-weighted Hilbert-Schmidt Independence Criterion (HSIC) to reduce the redundancy. Moreover, it adaptively learns view-wise importance for embedding to detect noisy views, and mitigates the label imbalance problem by focal loss. Experimental results on four real-world multi-view multi-label datasets demonstrate the effectiveness of the proposed method.

cs.LG↗

arXiv2023

Informative Data Selection with Uncertainty for Multi-modal Object Detection

Noise has always been nonnegligible trouble in object detection by creating confusion in model reasoning, thereby reducing the informativeness of the data. It can lead to inaccurate recognition due to the shift in the observed pattern, that requires a robust generalization of the models. To implement a general vision model, we need to develop deep learning models that can adaptively select valid information from multi-modal data. This is mainly based on two reasons. Multi-modal learning can break through the inherent defects of single-modal data, and adaptive information selection can reduce chaos in multi-modal data. To tackle this problem, we propose a universal uncertainty-aware multi-modal fusion model. It adopts a multi-pipeline loosely coupled architecture to combine the features and results from point clouds and images. To quantify the correlation in multi-modal information, we model the uncertainty, as the inverse of data information, in different modalities and embed it in the bounding box generation. In this way, our model reduces the randomness in fusion and generates reliable output. Moreover, we conducted a completed investigation on the KITTI 2D object detection dataset and its derived dirty data. Our fusion model is proven to resist severe noise interference like Gaussian, motion blur, and frost, with only slight degradation. The experiment results demonstrate the benefits of our adaptive fusion. Our analysis on the robustness of multi-modal fusion will provide further insights for future research.

cs.CV↗

arXiv2023

Modulated non-collinear magnetic structure of (Co$_{0.97}$$^{57}$Fe$_{0.03}$)$_{4}$Nb$_{2}$O$_{9}$ as revealed by Mössbauer spectroscopy

In this work, we present detailed $^{57}$Fe Mössbauer spectroscopy investigations of (Co$_{0.97}$$^{57}$Fe$_{0.03}$)$_{4}$Nb$_{2}$O$_{9}$ compound to study its possible magnetic structures. We have shown that the previously reported magnetic structures can not satisfactorily describe our low temperature Mössbauer spectra. Therefore, in combination with theoretical calculations, we have proposed a modulated helicoidal magnetic structure that can be used to simulate the whole series of our low temperature Mössbauer spectra. Our results suggest that the combination of previously reported different magnetic structures are only approximations of the average magnetic structure from our modulated helicoidal model. We anticipate that the proposed modulated non-collinear magnetic structure might shed light on the understanding of the complex magnetoelectric effects observed in this system.

cond-mat.str-el↗

arXiv2023

Muon radiography experiments on the subway overburden structure detection

Muon radiography is an innovative and non-destructive technique for internal density structure imaging, based on measuring the attenuation of cosmic-ray muons after they penetrate the target. Due to the strong penetration ability of muons, the detection range of muon radiography can reach the order of hundreds of meters or even kilometers. Using a portable muon detector composed of plastic scintillators and silicon photomultipliers, we performed a short-duration(1h) flux scanning experiment of the overburden above the platform and tunnel of the Xiaoying West Road subway station under construction. With the observation direction facing up, the detector is placed on the north side of the track and moved eastward from the platform section inside the station to the tunnel section. The scanning length is 264m and a total of 21 locations are observed. By comparing the observed and predicted values of the muon survival ratio at different locations, the experiment accurately detects the jump in thickness at the interface of the platform section and tunnel section. Furthermore, unknown anomalies caused by random placed light brick piles and side passage mouth above the observation locations are detected and confirmed later. This experiment verifies the feasibility of using natural muons to quickly detect abnormal structures of the overburden of tunnel, and shows that muon radiography has broad application prospects in tunnel safety and other similar aspects.

physics.ins-det↗

arXiv2022

A Benchmark for Automatic Medical Consultation System: Frameworks, Tasks and Datasets

In recent years, interest has arisen in using machine learning to improve the efficiency of automatic medical consultation and enhance patient experience. In this article, we propose two frameworks to support automatic medical consultation, namely doctor-patient dialogue understanding and task-oriented interaction. We create a new large medical dialogue dataset with multi-level finegrained annotations and establish five independent tasks, including named entity recognition, dialogue act classification, symptom label inference, medical report generation and diagnosis-oriented dialogue policy. We report a set of benchmark results for each task, which shows the usability of the dataset and sets a baseline for future studies. Both code and data is available from https://github.com/lemuria-wchen/imcs21.

cs.CL↗