检索结果-内蒙古大学图书馆

EchoTrack: Auditory Referring Multi-Object Tracking for Autonomous Driving

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Lin, Jiacheng Chen, Jiajun Peng, Kunyu He, Xuan Li, Zhiyong Stiefelhagen, Rainer Yang, Kailun The College of Computer Science and Electronic Engineering Hunan University Changsha410082 China The School of Robotics Hunan University Changsha410012 China The National Engineering Research Center of Robot Visual Perception and Control Technology Hunan University Changsha410082 China The Institute for Robotics and An-Thropomatics Karlsruhe Institute of Technology Karlsruhe76131 Germany

This paper introduces the task of Auditory Referring Multi-Object Tracking (AR-MOT), which dynamically tracks specific objects in a video sequence based on audio expressions and appears as a challenging problem in autonomous driving. Due to the lack of semantic modeling capacity in audio and video, existing works have mainly focused on text-based multiobject tracking, which often comes at the cost of tracking quality, interaction efficiency, and even the safety of assistance systems, limiting the application of such methods in autonomous driving. In this paper, we delve into the problem of AR-MOT from the perspective of audio-video fusion and audio-video tracking. We put forward EchoTrack, an end-to-end AR-MOT framework with dual-stream vision transformers. The dual streams are intertwined with our Bidirectional Frequency-domain Cross-attention Fusion Module (Bi-FCFM), which bidirectionally fuses audio and video features from both frequency- and spatiotemporal domains. Moreover, we propose the Audio-visual Contrastive Tracking Learning (ACTL) regime to extract homogeneous semantic features between expressions and visual objects by learning homogeneous features between different audio and video objects effectively. Aside from the architectural design, we establish the first set of large-scale AR-MOT benchmarks, including Echo-KITTI, Echo-KITTI+, and Echo-BDD. Extensive experiments on the established benchmarks demonstrate the effectiveness of the proposed EchoTrack and its components. The source code and datasets are available at https://***/lab206/EchoTrack. Copyright © 2024, The Authors. All rights reserved.

关键词： Semantics

Fast Image Processing Approach for the Reliable Quantification of Intracranial Hematoma Volume in Brain CT Images

学校读者我要写书评

暂无评论

Fast Image Processing Approach for the Reliable Quantificati...

control Instrumentation System Conference (CISCON)

作者： Vidhya V. Anjan Gudigar Raghavendra U. Akanksha Shetty Lekha Shree B. T. Parvathy Nair K. Ajay Hegde Girish R. Menon Computer Science and Engineering Manipal Institute of Technology Manipal Academy of Higher Education Manipal India Instrumentation and Control Engineering Manipal Institute of Technology Manipal Academy of Higher Education Manipal India Department of Neurosurgery Manipal Hospitals Karnataka India Department of Neurosurgery Kasturba Medical College Manipal Academy of Higher Education Manipal India

ISBN: (数字)9798350375480

ISBN: (纸本)9798350375497

Traumatic brain injury (TBI) is a neurological condition that can cause severe physical and mental disabilities or death. One of the serious consequences of TBI is intracranial hematoma that can be diagnosed through early and proper analysis of the brain CT images. Proper estimation of hematoma volume is essential for envisaging the 30-day outcome. The Tada formula is the gold standard used by radiologists to quantify the hematoma volume in CT images which will provide accurate results only for symmetrical hematoma configurations. Manual volume assessment is further subjective and time-consuming, especially in real-time acute clinical care set-up. Hence a computer-aided scheme that uses image processing techniques for rapid, precise, and reliable assessment of hematoma volume is developed. The CT images are pre-processed using histogram equalization and gray matter is further extracted using thresholding and morphological operations. Then the hematoma regions are extracted using thresholding and the 3D hematoma volume for the entire CT scan is computed by multiplying the image width and the total area of hematoma in all CT images. The obtained hematoma volumes are clinically validated by radiologist and the proposed approach achieved an average accuracy of 89.51%. A 3D modeling of the hematoma regions in a CT scan was further generated to provide an extensive visual aid for improved hematoma interpretation and strategic analysis. The entire process of automated hematoma volume computation and generation of 3D volume in a CT scan took around 13 seconds using standalone system. The outcomes of the experiment show that the proposed method can be used as a medical support system for rapid and accurate diagnosis of hematoma.

关键词： Visualization Three-dimensional displays Accuracy Thresholding (Imaging) Computed tomography Estimation Real-time systems Reliability Brain injuries Morphological operations

AdvDiffuser: Generating Adversarial Safety-Critical Driving Scenarios via Guided Diffusion

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Xie, Yuting Guo, Xianda Wang, Cong Liu, Kunhua Chen, Long School of Computer Science and Engineering Sun Yat-Sen University China School of Computer Science Wuhan University China State Key Laboratory for Management and Control of Complex Systems The Institute of Automation Chinese Academy of Sciences China School of Mechanical and Automotive Engineering Qingdao University of Technology China

Safety-critical scenarios are infrequent in natural driving environments but hold significant importance for the training and testing of autonomous driving systems. The prevailing approach involves generating safety-critical scenarios automatically in simulation by introducing adversarial adjustments to natural environments. These adjustments are often tailored to specific tested systems, thereby disregarding their transferability across different systems. In this paper, we propose AdvDiffuser, an adversarial framework for generating safety-critical driving scenarios through guided diffusion. By incorporating a diffusion model to capture plausible collective behaviors of background vehicles and a lightweight guide model to effectively handle adversarial scenarios, AdvDiffuser facilitates transferability. Experimental results on the nuScenes dataset demonstrate that AdvDiffuser, trained on offline driving logs, can be applied to various tested systems with minimal warm-up episode data and outperform other existing methods in terms of realism, diversity, and adversarial performance. © 2024, CC BY.

关键词： Automatic guided vehicles

Critical Test Cases Generalization for Autonomous Driving Object Detection Algorithms

学校读者我要写书评

暂无评论

Critical Test Cases Generalization for Autonomous Driving Ob...

IEEE Symposium on Intelligent Vehicle

作者： Zhengmin Jiang Jia Liu Ming Sang Huiyun Li Yi Pan Shenzhen Institute of Advanced Technology Chinese Academy of Sciences Shenzhen China University of Chinese Academy of Sciences (UCAS) Beijing China Guangdong-Hong Kong-Macao Joint Laboratory of Human-Machine Intelligence-Synergy Systems Dean of Computer Science and Control Engineering Shenzhen Institute of Advanced Technology Chinese Academy of Sciences Shenzhen China

ISBN: (数字)9798350348811

ISBN: (纸本)9798350348828

Visual-based object detection has become a crucial component in the realm of autonomous vehicles. However, conducting reliable testing for such systems remains unresolved. In this paper, we advocate for the application of causal inference to investigate the pivotal environmental factors influencing detection accuracy. Through the integration of diffusion models, we address the specialized conditional generalization of hazardous testing images. Our approach involves the construction of observational data to attribute key factors and fine-tune the diffusion model. Additionally, we introduce an optimal prompt words search method that strikes a balance between test coverage and level of challenge. Subsequently, leveraging these optimal prompts, we propose a cost-effective testing image generation through both "Text2Scene" and "Image2Scene" fashions. The experimental results indicate that, on the generalized dataset, the performance of object detection algorithms is the poorest, with the average detection accuracy decreasing from 0.81 to 0.285. Moreover, retraining object detection models on our generalized critical test cases can ultimately enhance algorithm performance, achieving a median accuracy improvement of up to 8.13%. Overall, our research proposes a novel approach to generalize test cases, thereby contributing to the advancement and deployment of safer autonomous vehicles.

关键词： Training Accuracy Search methods Text to image Object detection Diffusion models Inference algorithms

Dynamic Neural Potential Field: Online Trajectory Optimization in Presence of Moving Obstacles

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Staroverov, Aleksey Alhaddad, Muhammad Narendra, Aditya Mironov, Konstantin Panov, Aleksandr The Artificial Intelligence Research Institute Moscow105064 Russia Federal Research Center "Computer Science and Control Moscow117312 Russia Center of Cognitive Modeling Moscow Institute of Physics and Technology Dolgoprudny141701 Russia Faculty of Electrical and Electronic Engineering University of Aleppo Syria Ufa University of Science and Technology Ufa450000 Russia

We address a task of local trajectory planning for the mobile robot in the presence of static and dynamic obstacles. Local trajectory is obtained as a numerical solution of the Model Predictive control (MPC) problem. Collision avoidance may be provided by adding repulsive potential of the obstacles to the cost function of MPC. We develop an approach, where repulsive potential is estimated by the neural model. We propose and explore three possible strategies of handling dynamic obstacles. First, environment with dynamic obstacles is considered as a sequence of static environments. Second, the neural model predict a sequence of repulsive potential at once. Third, the neural model predict future repulsive potential step by step in autoregressive mode. We implement these strategies and compare it with CIAO* and MPPI using BenchMR framework. First two strategies showed higher performance than CIAO* and MPPI while preserving safety constraints. The third strategy was a bit slower, however it still satisfy time limits. We deploy our approach on Husky UGV mobile platform, which move through the office corridors under proposed MPC local trajectory planner. The code and trained models are available at https://***/CognitiveAISystems/ Dynamic-Neural-Potential-Field. © 2024, CC BY.

关键词： Cost functions

Annette: Accurate neural network execution time estimation with stacked models

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Wess, Matthias Ivanov, Matvey Nookala, Anvesh Wendt, Alexander Unger, Christoph Jantsch, Axel TU Wien Institute for Computer Technology TU Wien Institute of Automation and Control

With new accelerator hardware for Deep Neural Networks (DNNs), the computing power for Artificial Intelligence (AI) applications has increased rapidly. However, as DNN algorithms become more complex and optimized for specific applications, latency requirements remain challenging, and it is critical to find the optimal points in the design space. To decouple the architectural search from the target hardware, we propose a time estimation framework that allows for modeling the inference latency of DNNs on hardware accelerators based on mapping and layer-wise estimation models. The proposed methodology extracts a set of models from micro-kernel and multi-layer benchmarks and generates a stacked model for mapping and network execution time estimation. We compare estimation accuracy and fidelity of the generated mixed models, statistical models with the roofline model, and a refined roofline model for evaluation. We test the mixed models on the ZCU102 SoC board with Xilinx Deep Neural Network Development Kit (DNNDK) and Intel Neural Compute Stick 2 (NCS2) on a set of 12 state-of-the-art neural networks. It shows an average estimation error of 3.47% for the DNNDK and 7.44% for the NCS2, outperforming the statistical and analytical layer models for almost all selected networks. For a randomly selected subset of 34 networks of the NASBench dataset, the mixed model reaches fidelity of 0.988 in Spearman's ρ rank correlation coefficient metric. The code of ANNETTE is publicly available1 Copyright © 2021, The Authors. All rights reserved.

关键词： Deep neural networks

Att-TCN: An Advertising Attribution Method Based on Temporal Convolutional Network with Attention Mechanism 8

学校读者我要写书评

暂无评论

Att-TCN: An Advertising Attribution Method Based on Temporal...

8th IEEE International Conference on computer and Communications, ICCC 2022

作者： Xie, Bowen Sun, Yi Lin, Zhaowen Ma, Baoquan Beijing University of Posts and Telecommunications Beijing China National Engineering Laboratory for Mobile Network Technologies Beijing China National Engineering Laboratory for Industry Control System Information Security Technology Beijing China National Computer System Engineering Research Institute of China Beijing China

ISBN: (纸本)9781665450515

Advertising attribution is an effective method to measure the effect of various advertisements in the era of Internet advertising. It enables scientific tracking back and refined operation of advertising by quantitatively evaluating the contribution of different advertising channels to reach the target. It is challenging to predict conversion results based on users' journey data accurately, reflect the impact of ads on user conversions, and appropriately assign attribution values to advertising channels. To achieve the above objectives, we propose an advertising attribution method based on a temporal convolutional network with an attention mechanism (Att-TCN). By mining the hidden information in time and attribution channels using the attention mechanism, our proposed method can improve the accuracy of conversion prediction and thus more rationally assign attribution values across channels. Experimental validation on a real dataset obtained from an operational commercial advertising monitoring company shows that our proposed attribution method has better accuracy than other state-of-the-art attribution methods and gives an open-ended approach to attribution value assignment. © 2022 IEEE.

关键词： Marketing

A Comparative Study of Machine Learning Approaches for Autism Detection in Children from Imaging Data

学校读者我要写书评

暂无评论

A Comparative Study of Machine Learning Approaches for Autis...

2022 International Conference of Yearly Reports on Informatics, Mathematics, and Engineering, ICYRIME 2022

作者： Ponzi, Valerio Russo, Samuele Wajda, Agata Napoli, Christian Department of Computer Control and Management Engineering Sapienza University of Rome Via Ariosto 25 Roma00185 Italy Department of Psychology Sapienza University of Rome Via dei Marsi 78 Roma00185 Italy Institute of Energy and Fuel Processing Technology Zabrze41-803 Poland Institute for Systems Analysis and Computer Science Italian National Research Council Via dei Taurini 19 Roma00185 Italy

Autism Spectrum Disorder (ASD) is a neurodevelopmental condition that affects language, communication, cognitive, and social skills. Early detection of ASD in children is crucial for effective intervention, and machine learning techniques have emerged as promising tools to improve the accuracy and efficiency of detection. This paper presents a range of Machine Learning approaches that have been applied to identify individuals with ASD, with a particular focus on children, using images as input data. The results of these studies demonstrate the potential for Machine Learning to aid in the early detection and diagnosis of ASD in children, which can lead to better outcomes for individuals with this condition. © 2022 Copyright for this paper by its authors.

关键词： Machine learning

Rethinking Guidance Information to Utilize Unlabeled Samples: A Label Encoding Perspective

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Zhang, Yulong Yao, Yuan Chen, Shuhao Jin, Pengrong Zhang, Yu Jin, Jian Lu, Jiangang Department of Computer Science and Engineering Southern University of Science and Technology Shenzhen China College of Control Science and Engineering Zhejiang University Hangzhou China Beijing Teleinfo Technology Company Ltd. China Academy of Information and Communications Technology Beijing China Research Institute of Industrial Internet of Things China Academy of Information and Communications Technology Beijing China

Empirical Risk Minimization (ERM) is fragile in scenarios with insufficient labeled samples. A vanilla extension of ERM to unlabeled samples is Entropy Minimization (EntMin), which employs the soft-labels of unlabeled samples to guide their learning. However, EntMin emphasizes prediction discriminability while neglecting prediction diversity. To alleviate this issue, in this paper, we rethink the guidance information to utilize unlabeled samples. By analyzing the learning objective of ERM, we find that the guidance information for labeled samples in a specific category is the corresponding label encoding. Inspired by this finding, we propose a Label-Encoding Risk Minimization (LERM). It first estimates the label encodings through prediction means of unlabeled samples and then aligns them with their corresponding ground-truth label encodings. As a result, the LERM ensures both prediction discriminability and diversity, and it can be integrated into existing methods as a plugin. Theoretically, we analyze the relationships between LERM and ERM as well as EntMin. Empirically, we verify the superiority of the LERM under several label insufficient scenarios. The codes are available at https://***/zhangyl660/LERM. Copyright © 2024, The Authors. All rights reserved.

关键词： Forecasting