检索结果-内蒙古大学图书馆

Towards a Unified Transformer-based Framework for Scene Graph Generation and Human-object Interaction Detection

学校读者我要写书评

暂无评论

arXiv 2023年

作者： He, Tao Gao, Lianli Song, Jingkuan Li, Yuan-Fang Laboratory of Intelligent Collaborative Computing University of Electronic Science and Technology of China China Department of Data Science and AI Faculty of Information Technology Monash University Australia Center for Future Media University of Electronic Science and Technology of China China

Scene graph generation (SGG) and human-object interaction (HOI) detection are two important visual tasks aiming at localising and recognising relationships between objects, and interactions between humans and objects, respectively. Prevailing works treat these tasks as distinct tasks, leading to the development of task-specific models tailored to individual datasets. However, we posit that the presence of visual relationships can furnish crucial contextual and intricate relational cues that significantly augment the inference of human-object interactions. This motivates us to think if there is a natural intrinsic relationship between the two tasks, where scene graphs can serve as a source for inferring human-object interactions. In light of this, we introduce SG2HOI+, a unified one-step model based on the Transformer architecture. Our approach employs two interactive hierarchical Transformers to seamlessly unify the tasks of SGG and HOI detection. Concretely, we initiate a relation Transformer tasked with generating relation triples from a suite of visual features. Subsequently, we employ another transformer-based decoder to predict human-object interactions based on the generated relation triples. A comprehensive series of experiments conducted across established benchmark datasets including Visual Genome, V-COCO, and HICO-DET demonstrates the compelling performance of our SG2HOI+ model in comparison to prevalent one-stage SGG models. Remarkably, our approach achieves competitive performance when compared to state-of-the-art HOI methods. Additionally, we observe that our SG2HOI+ jointly trained on both SGG and HOI tasks in an end-to-end manner yields substantial improvements for both tasks compared to individualized training paradigms. © 2023, CC BY.

关键词： Benchmarking

A Survey on Fairness-Aware Recommender Systems

学校读者我要写书评

暂无评论

SSRN

SSRN 2023年

作者： Jin, Di Wang, Luzhi Zhang, He Zheng, Yizhen Ding, Weiping Xia, Feng Pan, Shirui College of Intelligence and Computing Tianjin University China Department of Data Science and AI Faculty of IT Monash University Australia School of Information Science and Technology Nantong University China School of Computing Technologies RMIT University Australia School of Information and Communication Technology Griffith University Australia

As information filtering services, recommender systems have extremely enriched our daily life by providing personalized suggestions and facilitating people in decision-making, which makes them vital and indispensable to human society in the information era. However, as people become more dependent on them, recent studies show that recommender systems potentially own unintentional impacts on society and individuals because of their unfairness (e.g., gender discrimination in job recommendations). To develop trustworthy services, it is crucial to devise fairness-aware recommender systems that can mitigate these bias issues. In this survey, we summarise existing methodologies and practices of fairness in recommender systems. Firstly, we present concepts of fairness in different recommendation scenarios, comprehensively categorize current advances, and introduce typical methods to promote fairness in different stages of recommender systems. Next, after introducing datasets and evaluation metrics applied to assess the fairness of recommender systems, we will delve into the significant influence that fairness-aware recommender systems exert on real-world industrial applications. Subsequently, we highlight the connection between fairness and other principles of trustworthy recommender systems, aiming to consider trustworthiness principles holistically while advocating for fairness. Finally, we summarize this review, spotlighting promising opportunities in comprehending concepts, frameworks, the balance between accuracy and fairness, and the ties with trustworthiness, with the ultimate goal of fostering the development of fairness-aware recommender systems. © 2023, The Authors. All rights reserved.

关键词： Recommender systems

Sparse graph counting and Kelley–Meka bounds for binary systems

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Filmus, Yuval Hatami, Hamed Hosseini, Kaave Kelman, Esty The Henry & Marilyn Taub Faculty of Computer Science Faculty of Mathematics Technion School of Computer Science McGill University Canada Department of Computer Science University of Rochester United States Department of Computer Science Department of Computing & Data Sciences Boston University CSAIL Massachusetts Institute of Technology United States

In a recent breakthrough, Kelley and Meka (FOCS 2023) obtained a strong upper bound on the density of sets of integers without non-trivial three-term arithmetic progressions. In this work, we extend their result, establishing similar bounds for all linear patterns defined by binary systems of linear forms, where "binary" indicates that every linear form depends on exactly two variables. Prior to our work, no strong bounds were known for such systems even in the finite field model setting. A key ingredient in our proof is a graph counting lemma. The classical graph counting lemma, developed by Thomason (Random Graphs 1985) and Chung, Graham, and Wilson (Combinatorica 1989), is a fundamental tool in combinatorics. For a fixed graph H, it states that the number of copies of H in a pseudorandom graph G is similar to the number of copies of H in a purely random graph with the same edge density as G. However, this lemma is only non-trivial when G is a dense graph. In this work, we prove a graph counting lemma that is also effective when G is sparse. Moreover, our lemma is well-suited for density increment arguments in additive number theory. As an immediate application, we obtain a strong bound for the Turán problem in abelian Cayley sum graphs: let Γ be a finite abelian group with odd order. If a Cayley sum graph on Γ does not contain any r-clique as a subgraph, it must have at most 2−Ωr(log1/16 |Γ|) · |Γ|2 edges. These results hinge on the technology developed by Kelley and Meka and the follow-up work by Kelley, Lovett, and Meka (STOC 2024). © 2023, CC BY.

关键词： Graph theory

Discrepancy and Thematic Bibliometric Analyses of the Remaining Limitations in Artificial Intelligence

学校读者我要写书评

暂无评论

Discrepancy and Thematic Bibliometric Analyses of the Remain...

2024 IEEE International Conference on computing, ICOCO 2024

作者： Mansor, Mahayaudin M. Ibrahim, Nurain Radi, Noor Fadhilah Ahmad Rahim, Nadirah Abdul Yahaya, Syarul Heiry Bakar, Mohd Aftar Abu Zakaria, Roslinazairimah School of Mathematical Sciences College of Computing Informatics and Mathematics Universiti Teknologi MARA Malaysia Institute of Big Data Analytics and Artificial Intelligence Universiti Teknologi MARA Malaysia Department of Electrical and Computer Engineering Kuliyyah of Engineering International Islamic University Malaysia Department of Science Computer and Mathematics Universiti Teknologi MARA Penang Campus Malaysia Department of Mathematical Sciences Faculty of Science & Technology Universiti Kebangsaan Malaysia Centre for Mathematical Sciences Universiti Malaysia Pahang Al-Sultan Abdullah Malaysia

ISBN: (纸本)9798331530303

As artificial intelligence (AI) advances, it is essential to continuously comprehend its limitations to optimise the integration of AI into autonomous systems that empower humans. The first objective of this study is to highlight the key capabilities of both humans and AI, focusing on the differences and identifying limitations of AI from the literature. Through discrepancy analysis, this study uses a Venn diagram to visualise the remaining limitations of AI into five key domains: (i) emotional intelligence, (ii) consciousness and awareness, (iii) creative imagination, (iv) communication, and (v) ethical decision-making. Furthermore, this study employs thematic bibliometric analysis to provide a more detailed examination of each AI limitation as the second objective. This study has identified underdeveloped and emerging research themes with potential for future development, such as emotion recognition, human-computer interaction, digital health, and situational awareness, which may require further research. Additionally, this study commends the ongoing efforts to harness AI’s computational power and algorithmic innovations to enhance AI’s overall performance and applicability. ©2024 IEEE.

关键词： Emotion Recognition

Local Information Guided Global Integration for Infrared Small Target Detection

学校读者我要写书评

暂无评论

Local Information Guided Global Integration for Infrared Sma...

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Qiang Li Qianchen Mao Wenjie Liu Jinbao Wang Wenmin Wang Bingshu Wang School of Software Northwestern Polytechnical University Xi’an China National Engineering Laboratory for Big Data System Computing Technology Shenzhen University Shenzhen China Faculty of Innovation Engineering Macau University of Science and Technology Macau China

Infrared small targets often exhibit small scale and weak semantic features, which makes it a great challenge to their detection. To address this situation, we propose a novel network for infrared small target detection that combines local details information and global contextual information. To preserve the local and high-frequency details present in infrared images, we introduce a High-frequency Aware Encoder. To extract contextual information from multi-scale feature maps, we propose a Multi-scale Context Learning Bottleneck that incorporates contextual information repeatedly and performs cross-level fusion, which enables the recognition of small targets based on their surroundings. Finally, a lightweight Transformer Decoder is employed to restore the feature map, while placing attention on the target pixels. Experimental results on the IRSTD-1k dataset demonstrate that our method outperforms other state-of-the-art approaches.

关键词：

Design of a W-band High-PAE Class A&AB Power Amplifier in 150nm GaAs Technology

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Lee, Jun Yan Wu, Duo Guo, Xuanrui Ariannejad, Mohammad Mahdi Bhuiyan, Mohammad Arif Sobhan Miraz, Mahdi H. School of Electronic Information and Electrical Engineering Shanghai Jiao Tong University Shanghai China Department of Electrical and Electronic Engineering Xiamen University Malaysia Sepang Malaysia Department of Microelectronics Delft University of Technology Delft Netherlands School of Computing and Data Science Xiamen University Malaysia Sepang Malaysia Faculty of Arts Science and Technology Wrexham University United Kingdom Faculty of Computing Engineering and Science University of South Wales Swansea United Kingdom

Nanometer scale power amplifiers (PA) at sub-THz suffer from severe parasitic effects that lead to experience limited maximum frequency and reduced power performance at the device transceiver front end. The integrated circuits researchers proposed different PA design architecture combinations at scaled down technologies to overcome these limitations. Although the designs meet the minimum requirements, the power added efficiency (PAE) of PA is still quite low. In this paper, a W-band single-ended common-source (CS) and cascode integrated 3-stage 2-way PA design is proposed. The design integrated different key design methodologies to mitigate the parasitic;such as combined Class AB and Class A stages for gain-boosting and efficiency enhancement, Wilkinson power combiner for higher output power, linearity, and bandwidth, and transmission line (TL)based wide band matching network for better inter-stage matching and compact size. The proposed PA design is validated using UMS 150-nm GaAs pHEMT using advanced design system (ADS) simulator. The results show that the proposed PA achieved a gain of 20.1 dB, an output power of 17.2 dBm, a PAE of 33 % and a 21 GHz bandwidth at 90 GHz Sub-THz band. The PA layout consumes only 5.66 × 2.51 mm2 die space including pads. Our proposed PA design will boost the research on sub-THz integrated circuits research and will smooth the wide spread adoption of 6G in near future. Copyright © 2024, The Authors. All rights reserved.

关键词： Power amplifiers

AGNN: Alternating Graph-Regularized Neural Networks to Alleviate Over-Smoothing

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Chen, Zhaoliang Wu, Zhihao Lin, Zhenghong Wang, Shiping Plant, Claudia Guo, Wenzhong The College of Computer and Data Science Fuzhou University Fuzhou350116 China The Fujian Provincial Key Laboratory of Network Computing and Intelligent Information Processing Fuzhou University Fuzhou350116 China The Faculty of Computer Science Austria The research network Data Science @ Uni Vienna University of Vienna Vienna1090 Austria

Graph Convolutional Network (GCN) with the powerful capacity to explore graph-structural data has gained noticeable success in recent years. Nonetheless, most of the existing GCN-based models suffer from the notorious over-smoothing issue, owing to which shallow networks are extensively adopted. This may be problematic for complex graph datasets because a deeper GCN should be beneficial to propagating information across remote neighbors. Recent works have devoted effort to addressing over-smoothing problems, including establishing residual connection structure or fusing predictions from multi-layer models. Because of the indistinguishable embeddings from deep layers, it is reasonable to generate more reliable predictions before conducting the combination of outputs from various layers. In light of this, we propose an Alternating Graph-regularized Neural Network (AGNN) composed of Graph Convolutional Layer (GCL) and Graph Embedding Layer (GEL). GEL is derived from the graph-regularized optimization containing Laplacian embedding term, which can alleviate the over-smoothing problem by periodic projection from the low-order feature space onto the high-order space. With more distinguishable features of distinct layers, an improved Adaboost strategy is utilized to aggregate outputs from each layer, which explores integrated embeddings of multi-hop neighbors. The proposed model is evaluated via a large number of experiments including performance comparison with some multi-layer or multi-order graph neural networks, which reveals the superior performance improvement of AGNN compared with state-of-the-art models. Copyright © 2023, The Authors. All rights reserved.

关键词： Embeddings

Drone Gesture Control using OpenCV and Tello

学校读者我要写书评

暂无评论

Drone Gesture Control using OpenCV and Tello

Conference on Information Communication Technology and Society (ICTAS)

作者： Nokuthula Khoza Pius Owolawi Vusumuzi Malele Department of Computer Systems Engineering Faculty of Information and Communication Technology Soshanguve Campus Tshwane University of Technology Pretoria South Africa Unit for Data Science and Computing School of Computer Science and Information Systems Vanderbijlpark Campus North-West University Vanderbijlpark South Africa

Tracking a person with an onboard camera is a very difficult and perhaps technically impossible if one camera is used. In this regard, real-life projects use a series of cameras to achieve the task. The advent of camera installed drone technology has made it possible for researchers and designers to resolve the tracking issues of a single or multiple stagnant devices. This paper presents a method that makes it possible for a drone to recognize hand gestures as control mechanism and follow the person who control it. Furthermore, the method assists in drone object detection using an integrated depth camera that provides video capturing. The OpenCV and Tello were used in this project. A stabilized depth video is for use with current person trackers like the OpenCV tracker. The estimated Tello pose deduced from vision to stabilize the depth image by warping it to a virtual-static camera. This method allows the Tello to obtain not only the tracked person's position and orientation but also their full body pose, which can then be used, for instance, to recognize hand gestures and alter the Tello's behavior. This work has a potential of contributing or enhancing cutting-edge applications such as hands-free photography and videography.

关键词：

Design of a W-band High-PAE Class A & AB Power Amplifier in 150nm GaAs Technology

学校读者我要写书评

暂无评论

SSRN

SSRN 2024年

Nanometer scale power amplifiers (PA) at sub-THz suffer from severe parasitic effects that lead to experience limited maximum frequency and reduced power performance at the device transceiver front end. The integrated circuits researchers proposed different PA design architecture combinations at scaled down technologies to overcome these limitations. Although the designs meet the minimum requirements, the power added efficiency (PAE) of PA is still quite low. In this paper, a W-band single-ended common-source (CS) and cascode integrated 3-stage 2-way PA design is proposed. The design integrated different key design methodologies to mitigate the parasitic;such as combined Class AB and Class A stages for gain-boosting and efficiency enhancement, Wilkinson power combiner for higher output power, linearity, and bandwidth, and transmission line (TL)-based wide band matching network for better inter-stage matching and compact size. The proposed PA design is validated using UMS 150-nm GaAs pHEMT using advanced design system (ADS) simulator. The results show that the proposed PA achieved a gain of 20.1 dB, an output power of 17.2 dBm, a PAE of 33 % and a 21 GHz bandwidth at 90 GHz Sub-THz band. The PA layout consumes only 5.66 × 2.51 mm2 die space including pads. Our proposed PA design will boost the research on sub-THz integrated circuits research and will smooth the wide spread adoption of 6G in near future. © 2024, The Authors. All rights reserved.

关键词： Power amplifiers