检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Lee, Saehyung Yoon, Seunghyun Bui, Trung Shi, Jing Yoon, Sungroh Department of Electrical and Computer Engineering Seoul National University Korea Republic of Adobe Research Interdisciplinary Program in Artificial Intelligence Seoul National University Korea Republic of

Multimodal large language models (MLLMs) excel at generating highly detailed captions but often produce hallucinations. Our analysis reveals that existing hallucination detection methods struggle with detailed captions. We attribute this to the increasing reliance of MLLMs on their generated text, rather than the input image, as the sequence length grows. To address this issue, we propose a multiagent approach that leverages LLM-MLLM collaboration to correct given captions. Additionally, we introduce an evaluation framework and a benchmark dataset to facilitate the systematic analysis of detailed captions. Our experiments demonstrate that our proposed evaluation method better aligns with human judgments of factuality than existing metrics and that existing approaches to improve the MLLM factuality may fall short in hyper-detailed image captioning tasks. In contrast, our proposed method significantly enhances the factual accuracy of captions, even improving those generated by GPT-4V. Finally, we highlight a limitation of VQA-centric benchmarking by demonstrating that an MLLM’s performance on VQA benchmarks may not correlate with its ability to generate detailed image captions. © 2024, CC BY.

关键词： Image enhancement

来源：评论

学校读者我要写书评

暂无评论

Hierarchical B-frame Video Coding Using Two-Layer CANF without Motion Coding

arXiv

引用

arXiv 2023年

作者： Alexandre, David Hang, Hsueh-Ming Peng, Wen-Hsiao Electrical Engineering and Computer Science International Graduate Program Institute of Electronics Department of Computer Science National Yang Ming Chiao Tung University Hsinchu Taiwan

Typical video compression systems consist of two main modules: motion coding and residual coding. This general architecture is adopted by classical coding schemes (such as international standards H.265 and H.266) and deep learning-based coding schemes. We propose a novel B-frame coding architecture based on two-layer Conditional Augmented Normalization Flows (CANF). It has the striking feature of not transmitting any motion information. Our proposed idea of video compression without motion coding offers a new direction for learned video coding. Our base layer is a low-resolution image compressor that replaces the full-resolution motion compressor. The low-resolution coded image is merged with the warped high-resolution images to generate a high-quality image as a conditioning signal for the enhancement-layer image coding in full resolution. One advantage of this architecture is significantly reduced computational complexity due to eliminating the motion information compressor. In addition, we adopt a skip-mode coding technique to reduce the transmitted latent samples. The rate-distortion performance of our scheme is slightly lower than that of the state-of-the-art learned B-frame coding scheme, B-CANF, but outperforms other learned B-frame coding schemes. However, compared to B-CANF, our scheme saves 45% of multiply–accumulate operations (MACs) for encoding and 27% of MACs for decoding. The code is available at https://***. Copyright © 2023, The Authors. All rights reserved.

关键词： Image coding

来源：评论

学校读者我要写书评

暂无评论

Calibrating Panoramic Depth Estimation for Practical Localization and Mapping

Calibrating Panoramic Depth Estimation for Practical Localiz...

引用

International Conference on computer Vision (ICCV)

作者： Junho Kim Eun Sun Lee Young Min Kim Dept. of Electrical and Computer Engineering Seoul National University Interdisciplinary Program in Artificial Intelligence and INMC Seoul National University

The absolute depth values of surrounding environments provide crucial cues for various assistive technologies, such as localization, navigation, and 3D structure estimation. We propose that accurate depth estimated from panoramic images can serve as a powerful and light-weight input for a wide range of downstream tasks requiring 3D information. While panoramic images can easily capture the surrounding context from commodity devices, the estimated depth shares the limitations of conventional image-based depth estimation; the performance deteriorates under large domain shifts and the absolute values are still ambiguous to infer from 2D observations. By taking advantage of the holistic view, we mitigate such effects in a self-supervised way and fine-tune the network with geometric consistency during the test phase. Specifically, we construct a 3D point cloud from the current depth prediction and project the point cloud at various viewpoints or apply stretches on the current input image to generate synthetic panoramas. Then we minimize the discrepancy of the 3D structure estimated from synthetic images without collecting additional data. We empirically evaluate our method in robot navigation and map-free localization where our method shows large performance enhancements. Our calibration method can therefore widen the applicability under various external conditions, serving as a key component for practical panorama-based machine vision systems.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Safe CoR: A Dual-Expert Approach to Integrating Imitation Learning and Safe Reinforcement Learning Using Constraint Rewards

Safe CoR: A Dual-Expert Approach to Integrating Imitation Le...

引用

IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

作者： Hyeokjin Kwon Gunmin Lee Junseo Lee Songhwai Oh Interdisciplinary Program in Artificial Intelligence and ASRI Seoul National University Seoul Korea Department of Electrical and Computer Engineering and ASRI Seoul National University Seoul Korea

ISBN: (数字)9798350377705

ISBN: (纸本)9798350377712

In the realm of autonomous agents, ensuring safety and reliability in complex and dynamic environments remains a paramount challenge. Safe reinforcement learning addresses these concerns by introducing safety constraints, but still faces challenges in navigating intricate environments such as complex driving situations. To overcome these challenges, we present the safe constraint reward (Safe CoR) framework, a novel method that utilizes two types of expert demonstrations—reward expert demonstrations focusing on performance optimization and safe expert demonstrations prioritizing safety. By exploiting a constraint reward (CoR), our framework guides the agent to balance performance goals of reward sum with safety constraints. We test the proposed framework in diverse environments, including the safety gym, metadrive, and the real-world Jackal platform. Our proposed framework improves algorithm performance by 39% and reduces constraint violations by 88% on the real-world Jackal platform, highlighting its effectiveness. Through this innovative approach, we expect significant advancements in real-world performance, leading to transformative effects in the realm of safe and reliable autonomous agents.

关键词： Navigation Imitation learning Focusing Reinforcement learning Autonomous agents Safety Reliability Optimization Intelligent robots Faces

来源：评论

学校读者我要写书评

暂无评论

LDL: Line Distance Functions for Panoramic Localization

LDL: Line Distance Functions for Panoramic Localization

引用

International Conference on computer Vision (ICCV)

作者： Junho Kim Changwoon Choi Hojun Jang Young Min Kim Dept. of Electrical and Computer Engineering Seoul National University Interdisciplinary Program in Artificial Intelligence and INMC Seoul National University

We introduce LDL, a fast and robust algorithm that localizes a panorama to a 3D map using line segments. LDL focuses on the sparse structural information of lines in the scene, which is robust to illumination changes and can potentially enable efficient computation. While previous line-based localization approaches tend to sacrifice accuracy or computation time, our method effectively observes the holistic distribution of lines within panoramic images and 3D maps. Specifically, LDL matches the distribution of lines with 2D and 3D line distance functions, which are further decomposed along principal directions of lines to increase the expressiveness. The distance functions provide coarse pose estimates by comparing the distributional information, where the poses are further optimized using conventional local feature matching. As our pipeline solely leverages line geometry and local features, it does not require costly additional training of line-specific features or correspondence matching. Nevertheless, our method demonstrates robust performance on challenging scenarios including object layout changes, illumination shifts, and large-scale scenes, while exhibiting fast pose search terminating within a matter of milliseconds. We thus expect our method to serve as a practical solution for line-based localization, and complement the well-established point-based paradigm. The code for LDL is available through the following link: https://***/82magnolia/panoramic-localization.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Arabic text sentiment analysis and emotion classification using transformers

Arabic text sentiment analysis and emotion classification us...

引用

Electronics, Communications, and Computations (JAC-ECC), International Japan-Africa Conference on

作者： Nada Mossad Youssef Mohamed A Fares Ahmed B. Zaky Egypt-Japan University of Science and Technology (E-JUST) Alexandria Egypt The Department of Electrical Engineering The Computer Systems Engineering Program Faculty of Engineering at Shoubra Benha University

Sentiment analysis and emotion classification are two crucial components of natural language processing (NLP), which have been widely explored in recent years due to their broad applications. Sentiment analysis aims to identify the polarity of written texts, ranging from positive to negative. Meanwhile, emotion classification is focused on recognizing and categorizing the emotional states expressed in the text. To achieve a deeper understanding of sentiments and emotions, it's essential to utilize models like BERT transformers that can effectively interpret the context. The process begins with data preprocessing, including tokenization and noise removal, followed by fine-tuning techniques to adapt the BERT model to the proposed tasks. We employed the BERT model on four datasets obtained from various sources, including Twitter, news websites, and restaurant reviews, where each dataset represents a distinct Arabic dialect. Our proposed model outperforms commonly used techniques like LSTM and CNN, yielding superior results. Despite the progress made, there are still challenges to overcome, such as dealing with Arabic diacritics, the new Arabic Arabizi, which uses Latin characters, and handling Arabic idioms. Further research is required to address these challenges adequately.

关键词：

来源：评论

学校读者我要写书评

暂无评论

ReMP: Reusable Motion Prior for Multi-domain 3D Human Pose Estimation and Motion Inbetweening

arXiv

引用

arXiv 2024年

作者： Jang, Hojun Kim, Young Min Dept. of Electrical and Computer Engineering Seoul National University Korea Republic of Interdisciplinary Program in Artificial Intelligence and INMC Seoul National University Korea Republic of

We present Reusable Motion prior (ReMP), an effective motion prior that can accurately track the temporal evolution of motion in various downstream tasks. Inspired by the success of foundation models, we argue that a robust spatio-temporal motion prior can encapsulate underlying 3D dynamics applicable to various sensor modalities. We learn the rich motion prior from a sequence of complete parametric models of posed human body shape. Our prior can easily estimate poses in missing frames or noisy measurements despite significant occlusion by employing a temporal attention mechanism. More interestingly, our prior can guide the system with incomplete and challenging input measurements to quickly extract critical information to estimate the sequence of poses, significantly improving the training efficiency for mesh sequence recovery. ReMP consistently outperforms the baseline method on diverse and practical 3D motion data, including depth point clouds, LiDAR scans, and IMU sensor data. Project page is available in https://***/ReMP. Copyright © 2024, The Authors. All rights reserved.

关键词： Reusability

来源：评论

学校读者我要写书评

暂无评论

A Fingerprint Based Indoor Visible Light Positioning System for Tilted Receivers Equipped with a Single Photo Detector

A Fingerprint Based Indoor Visible Light Positioning System ...

引用

IEEE Conference on Wireless Communications and Networking

作者： Ibrahim M. Abou-Shehada Ali H. Muqaibel Wessam Mesbah Ki-Hong Park Mohamed-Slim Alouini Electrical Engineering Department King Fahd University of Petroleum and Minerals Dhahran Saudi Arabia Electrical Engineering Department Center for Communication Systems and Sensing King Fahd University of Petroleum and Minerals Dhahran Saudi Arabia Electrical and Computer Engineering Program King Abdullah University of Science and Technology Thuwal Saudi Arabia

ISBN: (数字)9798350303582

ISBN: (纸本)9798350303599

Indoor positioning systems (IPS) are gaining higher attention recently due to the increased demand for indoor location aware services. Visible light communication (VLC) is a promising technology to use for IPS. In particular, received signal strength (RSS) based visible light positioning (VLP) systems are gaining high attention due to their low complexity and cost, in addition to higher positioning accuracy compared to their radio frequency (RF) counterparts. One of the main challenges in RSS based VLP systems is encountered when the receiver (the target) is tilted and not placed in parallel with the transmitters (the anchors). RSS based trilateration techniques require a computationally expensive and time-consuming process to solve the nonlinear problem of tilted receivers. Fingerprint based systems generally provide high positioning accuracy with short positioning time, and maybe used to circumvent the need to deal with the high complexity associated with tilted receivers. However, the design of a fingerprinting VLP system for tilted receiver has not been explored yet as far as receivers with a single photodetector (PD) are concerned. In this work, a fingerprint based VLP system for tilted receivers using artificial neural networks (ANN) is proposed, where different types of input features for training the positioning algorithm are studied. We show that using the components of the normal vector to the PD's surface in addition to RSS values provides an excellent positioning accuracy with an average positioning error of 25.41 cm and a remarkably low average positioning time less than $\mathbf{5} \boldsymbol{\mu} \boldsymbol{s}$ . In addition, important research directions for future work are discussed.

关键词： Training Accuracy Receivers Artificial neural networks Fingerprint recognition Vectors Real-time systems

来源：评论

学校读者我要写书评

暂无评论

RN-SDEs: Limited-Angle CT Reconstruction with Residual Null-Space Diffusion Stochastic Differential Equations

arXiv

引用

arXiv 2024年

作者： Guo, Jiaqi Lopez-Tapia, Santiago Li, Wing Shun Wu, Yunnan Carignano, Marcelo Backman, Vadim Dravid, Vinayak P. Katsaggelos, Aggelos K. The Department of Electrical and Computer Engineering Northwestern University IL60201 United States The Applied Physics Program Northwestern University EvanstonIL60201 United States

Computed tomography is a widely used imaging modality with applications ranging from medical imaging to material analysis. One major challenge arises from the lack of scanning information at certain angles, leading to distorted CT images with artifacts. This results in an ill-posed problem known as the Limited Angle Computed Tomography (LACT) reconstruction problem. To address this problem, we propose Residual Null-Space Diffusion Stochastic Differential Equations (RN-SDEs), which are a variant of diffusion models that characterize the diffusion process with mean-reverting (MR) stochastic differential equations. To demonstrate the generalizability of RN-SDEs, our experiments are conducted on two different LACT datasets, i.e., ChromSTEM and C4KC-KiTS. Through extensive experiments, we show that by leveraging learned Mean-Reverting SDEs as a prior and emphasizing data consistency using Range-Null Space Decomposition (RNSD) based rectification, RN-SDEs can restore high-quality images from severe degradation and achieve state-of-the-art performance in most LACT tasks. Additionally, we present a quantitative comparison of computational complexity and runtime efficiency, highlighting the superior effectiveness of our proposed approach. Copyright © 2024, The Authors. All rights reserved.

关键词： Stochastic systems

来源：评论

学校读者我要写书评

暂无评论

ConvNeXt Model for Breast Cancer Image Classification

ConvNeXt Model for Breast Cancer Image Classification

引用

International Conference on Cybernetics and Intelligent System (ICORIS)

作者： Devin Setiawan Andrea Stevens Karnyoto Indo Intan Bens Pardamean Department of Electrical Engineering and Computer Science University of Kansas Lawrence United States Bioinformatics and Data Science Research Center Bina Nusantara University Jakarta Indonesia Department of Informatics Engineering Universitas Dipa Makassar Makassar Indonesia Computer Science Department BINUS Graduate Program - Master of Computer Science Program Bina Nusantara University Jakarta Indonesia

ISBN: (数字)9798331539603

ISBN: (纸本)9798331539610

Breast cancer is an occurrence of cancer that attacks breast tissue and is the most common cancer among women worldwide, affecting one in eight women. In this modern world, breast cancer image classification simplifies the process of analyzing, providing objective and accurate results. By leveraging machine learning algorithms and computer vision techniques, we developed breast cancer detection. The dataset is histopathology dataset from BreakHis and UNHAS Hospital. We chose the ConvNeXt-Tiny model then modified the classifier head as the proposed method. Before the dataset is processed by the model, we augment the images by applying random horizontal and vertical flips, adjustments to brightness, contrast, saturation, and hue using color jitter. The augmentation process simulates real-world variance and enhances the model's ability to generalize to unseen data. Our proposed model gained better performance (accuracy, F1-Score) results compared two other techniques to VGG16 and SVM. According to our experiments, the F1-Score for the ConvNeXt-Tiny model with classifier head modification is higher at 0.9516, than the gain for VGG16 at 0.9292, and the gain for the SVM at 0.83.

关键词： Support vector machines Accuracy Head Computational modeling Transfer learning Breast cancer Magnetic heads Data models Reliability Image classification

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：