检索结果-内蒙古大学图书馆

BEVHeight: A Robust Framework for Vision-based Roadside 3D Object Detection

学校读者我要写书评

暂无评论

BEVHeight: A Robust Framework for Vision-based Roadside 3D O...

Conference on Computer Vision and Pattern Recognition (CVPR)

作者： Lei Yang Kaicheng Yu Tao Tang Jun Li Kun Yuan Li Wang Xinyu Zhang Peng Chen State Key Laboratory of Automotive Safety and Energy Tsinghua University Autonomous Driving Lab Alibaba Group Shenzhen Campus Sun Yat-sen University Center for Machine Learning Research Peking University

While most recent autonomous driving system focuses on developing perception methods on ego-vehicle sensors, people tend to overlook an alternative approach to leverage intelligent roadside cameras to extend the perception ability beyond the visual range. We discover that the state-of-the-art vision-centric bird's eye view detection methods have inferior performances on roadside cameras. This is because these methods mainly focus on recovering the depth regarding the camera center, where the depth difference between the car and the ground quickly shrinks while the distance increases. In this paper, we propose a simple yet effective approach, dubbed BEVHeight, to address this issue. In essence, instead of predicting the pixel-wise depth, we regress the height to the ground to achieve a distance-agnostic formulation to ease the optimization process of camera-only perception methods. On popular 3D detection benchmarks of roadside cameras, our method surpasses all previous vision-centric methods by a significant margin. The code is available at https://***/ADlab-AutoDrive/BEVHeight.

关键词：

GOLLIC: learning GLOBAL CONTEXT BEYOND PATCHES FOR LOSSLESS HIGH-RESOLUTION IMAGE COMPRESSION

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Lan, Yuan Qin, Liang Sun, Zhaoyi Xiang, Yang Sun, Jie Theory Lab Huawei Hong Kong Research Center Hong Kong Department of Mathematics The Hong Kong University of Science and Technology Hong Kong Algorithms of Machine Learning and Autonomous Driving Research Lab HKUST Shenzhen-Hong Kong Collaborative Innovation Research Institute Hong Kong

Neural-network-based approaches recently emerged in the field of data compression and have already led to significant progress in image compression, especially in achieving a higher compression ratio. In the lossless image compression scenario, however, existing methods often struggle to learn a probability model of full-size high-resolution images due to the limitation of the computation source. The current strategy is to crop high-resolution images into multiple non-overlapping patches and process them independently. This strategy ignores long-term dependencies beyond patches, thus limiting modeling performance. To address this problem, we propose a hierarchical latent variable model with a global context to capture the long-term dependencies of high-resolution images. Besides the latent variable unique to each patch, we introduce shared latent variables between patches to construct the global context. The shared latent variables are extracted by a self-supervised clustering module inside the model’s encoder. This clustering module assigns each patch the confidence that it belongs to any cluster. Later, shared latent variables are learned according to latent variables of patches and their confidence, which reflects the similarity of patches in the same cluster and benefits the global context modeling. Experimental results show that our global context model improves compression ratio compared to the engineered codecs and deep learning models on three benchmark high-resolution image datasets, DIV2K, ***, and ***. Copyright © 2022, The Authors. All rights reserved.

关键词： Image compression

BEVHeight: A Robust Framework for Vision-based Roadside 3D Object Detection

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Yang, Lei Yu, Kaicheng Tang, Tao Li, Jun Yuan, Kun Wang, Li Zhang, Xinyu Chen, Peng State Key Laboratory of Automotive Safety and Energy Tsinghua University China Autonomous Driving Lab Alibaba Group China Shenzhen Campus Sun Yat-sen University China Center for Machine Learning Research Peking University China

While most recent autonomous driving system focuses on developing perception methods on ego-vehicle sensors, people tend to overlook an alternative approach to leverage intelligent roadside cameras to extend the perception ability beyond the visual range. We discover that the state-of-the-art vision-centric bird’s eye view detection methods have inferior performances on roadside cameras. This is because these methods mainly focus on recovering the depth regarding the camera center, where the depth difference between the car and the ground quickly shrinks while the distance increases. In this paper, we propose a simple yet effective approach, dubbed BEVHeight, to address this issue. In essence, instead of predicting the pixel-wise depth, we regress the height to the ground to achieve a distance-agnostic formulation to ease the optimization process of camera-only perception methods. On popular 3D detection benchmarks of roadside cameras, our method surpasses all previous vision-centric methods by a significant margin. The code is available at https://***/ ADlab-AutoDrive/BEVHeight. Copyright © 2023, The Authors. All rights reserved.

关键词： Cameras

Nearly Optimal VC-Dimension and Pseudo-Dimension Bounds for Deep Neural Network Derivatives

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Yang, Yahong Yang, Haizhao Xiang, Yang Department of Mathematics Hong Kong University of Science and Technology Clear Water Bay Hong Kong Department of Mathematics Department of Computer Science University of Maryland College Park College ParkMD United States Algorithms of Machine Learning and Autonomous Driving Research Lab HKUST Shenzhen-Hong Kong Collaborative Innovation Research Institute Futian Shenzhen China

This paper addresses the problem of nearly optimal Vapnik–Chervonenkis dimension (VC-dimension) and pseudo-dimension estimations of the derivative functions of deep neural networks (DNNs). Two important applications of these estimations include: 1) Establishing a nearly tight approximation result of DNNs in the Sobolev space;2) Characterizing the generalization error of machine learning methods with loss functions involving function derivatives. This theoretical investigation fills the gap of learning error estimations for a wide range of physics-informed machine learning models and applications including generative models, solving partial differential equations, operator learning, network compression, distillation, regularization, etc. Copyright © 2023, The Authors. All rights reserved.

关键词： Deep neural networks

ELASTIC INTERACTION ENERGY-INFORMED REAL-TIME TRAFFIC SCENE PERCEPTION

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Feng, Yaxin Lan, Yuan Zhang, Luchan Liu, Guoqing Xiang, Yang Department of Mathematics Hong Kong University of Science and Technology Clear Water Bay Hong Kong College of Mathematics and Statistics Shenzhen University Shenzhen China Shenzhen Youjia Innov Tech Co. Ltd. Shenzhen China Algorithms of Machine Learning and Autonomous Driving Research Lab HKUST Shenzhen-Hong Kong Collaborative Innovation Research Institute Futian Shenzhen China

Urban segmentation and lane detection are two important tasks for traffic scene perception. Accuracy and fast inference speed of visual perception are crucial for autonomous driving safety. Fine and complex geometric objects are the most challenging but important recognition targets in traffic scene, such as pedestrians, traffic signs and lanes. In this paper, a simple and efficient topology-aware energy loss function-based network training strategy named EIEGSeg is proposed. EIEGSeg is designed for multi-class segmentation on real-time traffic scene perception. To be specific, the convolutional neural network (CNN) extracts image features and produces multiple outputs, and the elastic interaction energy loss function (EIEL) drives the predictions moving toward the ground truth until they are completely overlapped. Our strategy performs well especially on fine-scale structure, i.e. small or irregularly shaped objects can be identified more accurately, and discontinuity issues on slender objects can be improved. We quantitatively and qualitatively analyze our method on three traffic datasets, including urban scene segmentation data Cityscapes and lane detection data TuSimple and CULane. Our results demonstrate that EIEGSeg consistently improves the performance, especially on real-time, lightweight networks that are better suited for autonomous driving. Copyright © 2023, The Authors. All rights reserved.

关键词： Digital storage

LARGE TRANSFORMERS ARE BETTER EEG LEARNERS

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Wang, Bingxin Fu, Xiaowen Lan, Yuan Zhang, Luchan Zheng, Wei Xiang, Yang Department of Mathematics The Hong Kong University of Science and Technology Clear Water Bay Hong Kong Algorithms of Machine Learning and Autonomous Driving Research Lab HKUST Shenzhen-Hong Kong Collaborative Innovation Research Institute Futian Shenzhen China College of Mathematics and Statistics Shenzhen University Shenzhen518060 China Shenzhen Youjia Innov Tech Co. Ltd. Shenzhen China

Pre-trained large transformer models have achieved remarkable performance in the fields of natural language processing and computer vision. However, the limited availability of public electroencephalogram (EEG) data presents a unique challenge for extending the success of these models to EEG-based tasks. To address this gap, we propose AdaCT, plug-and-play Adapters designed for Converting Time series data into spatio-temporal 2D pseudo-images or text forms. Essentially, AdaCT-I transforms multi-channel or lengthy single-channel time series data into spatio-temporal 2D pseudo-images for fine-tuning pre-trained vision transformers, while AdaCT-T converts short single-channel data into text for fine-tuning pre-trained language transformers. The proposed approach allows for seamless integration of pre-trained vision models and language models in time series decoding tasks, particularly in EEG data analysis. Experimental results on diverse benchmark datasets, including Epileptic Seizure Recognition, Sleep-EDF, and UCI HAR, demonstrate the superiority of AdaCT over baseline methods. Overall, we provide a promising transfer learning framework for leveraging the capabilities of pre-trained vision and language models in EEG-based tasks, thereby advancing the field of time series decoding and enhancing interpretability in EEG data analysis. Our code will be available at https://***/wangbxj1234/AdaCE. © 2023, CC BY.

关键词： Electroencephalography

DOSNET AS A NON-BLACK-BOX PDE SOLVER: WHEN DEEP learning MEETS OPERATOR SPLITTING

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Lan, Yuan Li, Zhen Sun, Jie Xiang, Yang Theory Lab Huawei Technologies Co. Ltd. Hong Kong Department of Mathematics The Hong Kong University of Science and Technology Clear Water Bay Hong Kong Algorithms of Machine Learning and Autonomous Driving Research Lab HKUST Shenzhen-Hong Kong Collaborative Innovation Research Institute Futian Shenzhen China

Deep neural networks (DNNs) recently emerged as a promising tool for analyzing and solving complex differential equations arising in science and engineering applications. Alternative to traditional numerical schemes, learning-based solvers utilize the representation power of DNNs to approximate the input-output relations in an automated manner. However, the lack of physics-in-the-loop often makes it difficult to construct a neural network solver that simultaneously achieves high accuracy, low computational burden, and interpretability. In this work, focusing on a class of evolutionary PDEs characterized by having decomposable operators, we show that the classical "operator splitting" numerical scheme of solving these equations can be exploited to design neural network architectures. This gives rise to a learning-based PDE solver, which we name Deep Operator-Splitting Network (DOSnet). Such non-black-box network design is constructed from the physical rules and operators governing the underlying dynamics contains learnable parameters, and is thus more flexible than the standard operator splitting scheme. Once trained, it enables the fast solution of the same type of PDEs. To validate the special structure inside DOSnet, we take the linear PDEs as the benchmark and give the mathematical explanation for the weight behavior. Furthermore, to demonstrate the advantages of our new AI-enhanced PDE solver, we train and validate it on several types of operator-decomposable differential equations. We also apply DOSnet to nonlinear Schrödinger equations (NLSE) which have important applications in the signal processing for modern optical fiber transmission systems, and experimental results show that our model has better accuracy and lower computational complexity than numerical schemes and the baseline DNNs. Copyright © 2022, The Authors. All rights reserved.

关键词： Deep neural networks

Adaptive and hybrid reduced order models to mitigate Kolmogorov barrier in a multiscale kinetic transport equation

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Jin, Tianyu Peng, Zhichao Xiang, Yang Department of Mathematics The Hong Kong University of Science and Technology Hong Kong Algorithms of Machine Learning and Autonomous Driving Research Lab HKUST Shenzhen-Hong Kong Collaborative Innovation Research Institute Shenzhen Futian China

In this work, we develop reduced order models (ROMs) to predict solutions to a multiscale kinetic transport equation with a diffusion limit under the parametric setting. When the underlying scattering effect is not sufficiently strong, the system governed by this equation exhibits transport-dominated behavior. Suffering from the Kolmogorov barrier for transport dominant problems, classical linear ROMs may become inefficient in this regime. To address this issue, we first develop a piecewise linear ROM by introducing a novel goal-oriented adaptive time partitioning strategy. To avoid local over-refinement or under-refinement, we propose an adaptive coarsening and refinement strategy that remains robust with various initial empirical partitions. Additionally, for problems where a local linear approximation is not sufficiently efficient, we further develop a hybrid ROM, which combines autoencoder-based nonlinear ROMs and piecewise linear ROMs. Compared to previous autoencoder-based ROMs, this hybridized method reduces the offline autoencoder’s training cost by only applying it to time intervals that are adaptively identified as the most challenging. Numerical experiments demonstrate that our proposed approaches successfully predict full-order solutions at unseen parameter values with both efficiency and accuracy. To the best of our knowledge, this is the first attempt to address the Kolmogorov barrier for multiscale kinetic transport problems with the coexistence of both transport- and diffusion-dominant behaviors. Copyright © 2025, The Authors. All rights reserved.

关键词： Piecewise linear techniques

QUANTIFYING TRAINING DIFFICULTY AND ACCELERATING CONVERGENCE IN NEURAL NETWORK-BASED PDE SOLVERS

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Chen, Chuqi Zhou, Qixuan Yang, Yahong Xiang, Yang Luo, Tao Department of Mathematics Hong Kong University of Science and Technology Clear Water Bay Hong Kong School of Mathematical Sciences Shanghai Jiao Tong University Shanghai China Department of Mathematics The Pennsylvania State University University Park State CollegePA United States Algorithms of Machine Learning and Autonomous Driving Research Lab HKUST Shenzhen-Hong Kong Collaborative Innovation Research Institute Futian Shenzhen China School of Mathematical Sciences Institute of Natural Sciences MOE-LSC Shanghai Jiao Tong University CMA-Shanghai Shanghai Artificial Intelligence Laboratory Shanghai China

Neural network-based methods have emerged as powerful tools for solving partial differential equations (PDEs) in scientific and engineering applications, particularly when handling complex domains or incorporating empirical data. These methods leverage neural networks as basis functions to approximate PDE solutions. However, training such networks can be challenging, often resulting in limited accuracy. In this paper, we investigate the training dynamics of neural network-based PDE solvers with a focus on the impact of initialization techniques. We assess training difficulty by analyzing the eigenvalue distribution of the kernel and apply the concept of effective rank to quantify this difficulty, where a larger effective rank correlates with faster convergence of the training error. Building upon this, we discover through theoretical analysis and numerical experiments that two initialization techniques, partition of unity (PoU) and variance scaling (VS), enhance the effective rank, thereby accelerating the convergence of training error. Furthermore, comprehensive experiments using popular PDE-solving frameworks, such as PINN, Deep Ritz, and the operator learning framework DeepOnet, confirm that these initialization techniques consistently speed up convergence, in line with our theoretical findings. Copyright © 2024, The Authors. All rights reserved.

关键词： Eigenvalues and eigenfunctions