检索结果-内蒙古大学图书馆

International Russian Smart Industry Conference (SmartIndustryCon)

作者： Diana Bogdanova Azamat Giniiatullin Jaspreet Singh Batth Ufa University of Science and Technology Institute of Computer Science Mathematics and Robotics Ufa Russia Chandigarh University Computer Science and Engineering Chandigarh India

Content analysis of pilot speech is a key tool for improving flight safety as it allows to identify potential problems in communication between pilots, dispatchers and crew. The development of efficient algorithms for analyzing pilot speech contributes to improving communication skills, preventing incidents and improving overall airline safety. The research focuses on the possibility of applying machine learning techniques and deep neural networks to analyze speech data, which opens new perspectives for information processing and interpretation. The results of the study emphasize the interdisciplinary nature of the topic and suggest new methodological approaches, including the use of fuzzy set theory and fuzzy logic systems to handle uncertainty in speech data, which in turn helps to improve communication performance in complex and stressful flight conditions. The finalization of the work includes the evaluation of the developed methods and approaches in terms of their applicability and impact on improving flight safety, incorporating model training results and addressing testing errors.

关键词：

来源：评论

学校读者我要写书评

暂无评论

ReLU-QP: A GPU-Accelerated Quadratic Programming Solver for Model-Predictive Control

ReLU-QP: A GPU-Accelerated Quadratic Programming Solver for ...

引用

IEEE International Conference on robotics and Automation (ICRA)

作者： Arun L. Bishop John Z. Zhang Swaminathan Gurumurthy Kevin Tracy Zachary Manchester Robotics Institute School of Computer Science Carnegie Mellon University Pittsburgh PA USA

ISBN: (数字)9798350384574

ISBN: (纸本)9798350384581

We present ReLU-QP, a GPU-accelerated solver for quadratic programs (QPs) that is capable of solving high-dimensional control problems at real-time rates. ReLU-QP is derived by exactly reformulating the Alternating Direction Method of Multipliers (ADMM) algorithm for solving QPs as a deep, weight-tied neural network with rectified linear unit (ReLU) activations. This reformulation enables the deployment of ReLU-QP on GPUs using standard machine-learning toolboxes. We evaluate the performance of ReLU-QP across three model-predictive control (MPC) benchmarks: stabilizing random linear dynamical systems with control limits, balancing an Atlas humanoid robot on a single foot, and performing a whole-body pick-up motion on a quadruped equipped with a six-degree-of-freedom arm. These benchmarks indicate that ReLU-QP is competitive with state-of-the-art CPU-based solvers for small-to-medium-scale problems and offers order-of-magnitude speed improvements for larger-scale problems.

关键词： Neural networks Machine learning Benchmark testing Manipulators Real-time systems Quadrupedal robots Quadratic programming

来源：评论

学校读者我要写书评

暂无评论

D3GU: Multi-target Active Domain Adaptation via Enhancing Domain Alignment

D3GU: Multi-target Active Domain Adaptation via Enhancing Do...

引用

IEEE Workshop on Applications of computer Vision (WACV)

作者： Lin Zhang Linghan Xu Saman Motamed Shayok Chakraborty Fernando De la Torre Robotics Institute Carnegie Mellon University Department of Computer Science Florida State University

Unsupervised domain adaptation (UDA) for image classification has made remarkable progress in transferring classification knowledge from a labeled source domain to an unlabeled target domain, thanks to effective domain alignment techniques. Recently, in order to further improve performance on a target domain, many Single-Target Active Domain Adaptation (ST-ADA) methods have been proposed to identify and annotate the salient and exemplar target samples. However, it requires one model to be trained and deployed for each target domain and the domain label associated with each test sample. This largely restricts its application in the ubiquitous scenarios with multiple target domains. Therefore, we propose a Multi-Target Active Domain Adaptation (MT-ADA) framework for image classification, named D 3 GU, to simultaneously align different domains and actively select samples from them for annotation. This is the first research effort in this field to our best knowledge. D 3 GU applies Decomposed Domain Discrimination (D 3 ) during training to achieve both source-target and target-target domain alignments. Then during active sampling, a Gradient Utility (GU) score is designed to weight every unlabeled target image by its contribution towards classification and domain alignment tasks, and is further combined with KMeans clustering to form GU-KMeans for diverse image sampling. Extensive experiments on three benchmark datasets, Office31, OfficeHome, and DomainNet, have been conducted to validate consistently superior performance of D 3 GU for MT-ADA 1 .

关键词：

来源：评论

学校读者我要写书评

暂无评论

On the Faithfulness of Vision Transformer Explanations

On the Faithfulness of Vision Transformer Explanations

引用

Conference on computer Vision and Pattern Recognition (CVPR)

作者： Junyi Wu Weitai Kang Hao Tang Yuan Hong Yan Yan Department of Computer Science Illinois Institute of Technology USA Robotics Institute Carnegie Mellon University USA Department of Computer Science University of Connecticut USA

ISBN: (数字)9798350353006

ISBN: (纸本)9798350353013

To interpret Vision Transformers, post-hoc explanations assign salience scores to input pixels, providing human-understandable heatmaps. However, whether these inter-pretations reflect true rationales behind the model's output is still underexplored. To address this gap, we study the faithfulness criterion of explanations: the assigned salience scores should represent the influence of the corresponding input pixels on the model's predictions. To evaluate faithful-ness, we introduce Salience-guided Faithfulness Coefficient (SaCo), a novel evaluation metric leveraging essential in-formation of salience distribution. Specifically, we con-duct pair-wise comparisons among distinct pixel groups and then aggregate the differences in their salience scores, resulting in a coefficient that indicates the explanation's degree of faithfulness. Our explorations reveal that cur-rent metrics struggle to differentiate between advanced ex-planation methods and Random Attribution, thereby failing to capture the faithfulness property. In contrast, our pro-posed SaCo offers a reliable faithfulness measurement, es-tablishing a robust metric for interpretations. Furthermore, our SaCo demonstrates that the use of gradient and multi-layer aggregation can markedly enhance the faithfulness of attention-based explanation, shedding light on potential paths for advancing Vision Transformer explainability.

关键词： Measurement Heating systems computer vision Correlation Aggregates Predictive models Benchmark testing

来源：评论

学校读者我要写书评

暂无评论

Analog, Programmable Switched Capacitor FIR Filter Based on Rotator Architecture Implemented in CMOS Technology 26

Analog, Programmable Switched Capacitor FIR Filter Based on ...

引用

26th IEEE Signal Processing: Algorithms, Architectures, Arrangements, and Applications, SPA 2023

作者： Pawlowski, Pawel Dlugosz, Rafal Radkowski, Marek Watly, Mateusz Dabrowski, Adam Poznań University of Technology Division of Signal Processing and Electronic Systems Faculty of Control Robotics and Electrical Engineering Institute of Automatic Control and Robotics Poznań60-965 Poland Bydgoszcz University of Science and Technology Faculty of Telecommunication Computer Science and Electrical Engineering Bydgoszcz85-796 Poland

ISBN: (纸本)9798350304985

The paper presents project and its verification of a prototype integrated circuit containing an analog, programmable finite impulse response (FIR) filter, implemented in CMOS 350 nm technology. The structure of the filter is based on the switched capacitor technique. In circuits of this type, one of main challenges is an efficient implementation of filter coefficients, which result from several factors described in this work. When implementing such filters as programmable circuits, the values of their coefficients have to be limited to a selected range, i.e. a given resolution in bits. In the implemented prototype filter, the filter coefficients are represented by 6 bits in sign-magnitude notation, so they can take 63 different values only. In such filters, it is not possible to directly implement any frequency response of the filter. Each time, it is necessary to properly round the theoretical values of the coefficients so that they fit into the available range of discrete values resulting from the implementation. The authors of the work designed an algorithm that allows such matching. The paper also presents results of measurements of the prototype chip. © 2023 Division of Signal Processing and Electronic Systems, Poznan University of Technology (DSPES PUT).

关键词： FIR filters

来源：评论

学校读者我要写书评

暂无评论

LiLMaps: Learnable Implicit Language Maps

arXiv

引用

arXiv 2025年

作者： Kruzhkov, Evgenii Behnke, Sven Autonomous Intelligent Systems Computer Science Institute VI University of Bonn Germany Autonomous Intelligent Systems Computer Science Institute VI – Intelligent Systems and Robotics Center for Robotics and The Lamarr Institute for Machine Learning and Artificial Intelligence University of Bonn Germany

One of the current trends in robotics is to employ large language models (LLMs) to provide non-predefined command execution and natural human-robot interaction. It is useful to have an environment map together with its language representation, which can be further utilized by LLMs. Such a comprehensive scene representation enables numerous ways of interaction with the map for autonomously operating robots. In this work, we present an approach that enhances incremental implicit mapping through the integration of vision-language features. Specifically, we (i) propose a decoder optimization technique for implicit language maps which can be used when new objects appear on the scene, and (ii) address the problem of inconsistent vision-language predictions between different viewing positions. Our experiments demonstrate the effectiveness of LiLMaps and solid improvements in performance. © 2025, CC BY-NC-ND.

关键词： Visual languages

来源：评论

学校读者我要写书评

暂无评论

HairDiffusion: Vivid Multi-Colored Hair Editing via Latent Diffusion 38

HairDiffusion: Vivid Multi-Colored Hair Editing via Latent D...

引用

38th Conference on Neural Information Processing Systems, NeurIPS 2024

作者： Zeng, Yu Zhang, Yang Liu, Jiachen Shen, Linlin Deng, Kaijun He, Weizhao Wang, Jinbao Computer Vision Institute School of Computer Science & Software Engineering Shenzhen University China Shenzhen Institute of Artificial Intelligence and Robotics for Society China National Engineering Laboratory for Big Data System Computing Technology Shenzhen University China Guangdong Provincial Key Laboratory of Intelligent Information Processing China

Hair editing is a critical image synthesis task that aims to edit hair color and hairstyle using text descriptions or reference images, while preserving irrelevant attributes (e.g., identity, background, cloth). Many existing methods are based on StyleGAN to address this task. However, due to the limited spatial distribution of StyleGAN, it struggles with multiple hair color editing and facial preservation. Considering the advancements in diffusion models, we utilize Latent Diffusion Models (LDMs) for hairstyle editing. Our approach introduces Multi-stage Hairstyle Blend (MHB), effectively separating control of hair color and hairstyle in diffusion latent space. Additionally, we train a warping module to align the hair color with the target region. To further enhance multi-color hairstyle editing, we fine-tuned a CLIP model using a multi-color hairstyle dataset. Our method not only tackles the complexity of multi-color hairstyles but also addresses the challenge of preserving original colors during diffusion editing. Extensive experiments showcase the superiority of our method in editing multi-color hairstyles while preserving facial attributes given textual descriptions and reference images. © 2024 Neural information processing systems foundation. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Weisfeiler and Leman Go Loopy: A New Hierarchy for Graph Representational Learning 38

Weisfeiler and Leman Go Loopy: A New Hierarchy for Graph Rep...

引用

38th Conference on Neural Information Processing Systems, NeurIPS 2024

作者： Paolino, Raffaele Maskey, Sohir Welke, Pascal Kutyniok, Gitta Department of Mathematics LMU Munich Germany Germany Faculty of Computer Science TU Wien Austria Institute for Robotics and Mechatronics DLR-German Aerospace Center Germany Department of Physics and Technology University of Tromsø Norway

We introduce r-loopy Weisfeiler-Leman (r-`WL), a novel hierarchy of graph isomorphism tests and a corresponding GNN framework, r-`MPNN, that can count cycles up to length r+2. Most notably, we show that r-`WL can count homomorphisms of cactus graphs. This extends 1-WL, which can only count homomorphisms of trees and, in fact, we prove that r-`WL is incomparable to k-WL for any fixed k. We empirically validate the expressive and counting power of r-`MPNN on several synthetic datasets and demonstrate the scalability and strong performance on various real-world datasets, particularly on sparse graphs. Our code is available on GitHub. © 2024 Neural information processing systems foundation. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Exploiting Cross-Modal Cost Volume for Multi-sensor Depth Estimation 17th

Exploiting Cross-Modal Cost Volume for Multi-sensor Depth E...

引用

17th Asian Conference on computer Vision, ACCV 2024

作者： Kim, Janghyun Shin, Ukcheol Heo, Seokyong Park, Jinsun Department of Information Convergence Engineering Pusan National University Busan Korea Republic of Robotics Institute Carnegie Mellon University Pittsburgh United States School of Computer Science and Engineering Pusan National University Busan Korea Republic of

ISBN: (纸本)9789819609680

Single-modal depth estimation has shown steady improvement over the years. However, relying solely on a single imaging sensor such as RGB and near-infrared (NIR) cameras can result in unreliable and erroneous depth estimation, particularly in challenging lighting conditions such as low-light or sudden lighting change scenarios. Thereby, several approaches have leveraged multiple sensors for robust depth estimation. However, the effective fusion method that maximally utilizes multi-modal sensor information still requires further investigation. With this in mind, we propose a multi-modal cost volume fusion strategy with cross-modal attention, incorporating information from both cross-spectral and single-modality pairs. Our method initially constructs low-level cost volumes that consist of modality-specific (ı.e., single modality) and modality-invariant (ı.e., cross-spectral) volumes from multi-modal sensors. These cost volumes are then gradually fused using bidirectional cross-modal fusion and unidirectional LiDAR fusion to generate a multi-sensory cost volume. Furthermore, we introduce a straightforward domain gap reduction approach to learn modality-invariant features and depth refinement techniques through cost volume-guided propagation. Experimental results demonstrate that our method achieves SOTA (State-of-the-Art) performance under diverse environmental changes. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Depth perception

来源：评论

学校读者我要写书评

暂无评论

Token Transformation Matters: Towards Faithful Post-Hoc Explanation for Vision Transformer

Token Transformation Matters: Towards Faithful Post-Hoc Expl...

引用

Conference on computer Vision and Pattern Recognition (CVPR)

作者： Junyi Wu Bin Duan Weitai Kang Hao Tang Yan Yan Department of Computer Science Illinois Institute of Technology USA Robotics Institute Carnegie Mellon University USA

ISBN: (数字)9798350353006

ISBN: (纸本)9798350353013

While Transformers have rapidly gained popularity in various computer vision applications, post-hoc explanations of their internal mechanisms remain largely unexplored. Vision Transformers extract visual information by representing image regions as transformed tokens and integrating them via attention weights. However, existing post-hoc explanation methods merely consider these attention weights, neglecting crucial information from the transformed tokens, which fails to accurately illustrate the rationales behind the models' predictions. To incorporate the influence of token transformation into interpretation, we propose TokenTM, a novel post-hoc explanation method that utilizes our introduced measurement of token transformation effects. Specifically, we quantify token transformation effects by measuring changes in token lengths and correlations in their directions pre- and post-transformation. Moreover, we develop initialization and aggregation rules to integrate both attention weights and token transformation effects across all layers, capturing holistic token contributions throughout the model. Experimental results on segmentation and perturbation tests demonstrate the superiority of our proposed TokenTM compared to state-of-the-art Vision Transformer explanation methods.

关键词： computer vision Visualization Correlation Computational modeling Perturbation methods Predictive models Length measurement

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：