Article
Computer Science, Artificial Intelligence
Jonghyun Kim, Gen Li, Inyong Yun, Cheolkon Jung, Joongkyu Kim
Summary: This paper introduces a weakly-supervised temporal attention 3D network, TA3DNet, for human action recognition, which accelerates 3D convolutional neural networks by assigning different importance to each frame. The method achieves high performance on two challenging datasets and outperforms state-of-the-art methods for action recognition.
PATTERN RECOGNITION
(2021)
Article
Engineering, Electrical & Electronic
Liubing Jiang, Minyang Wu, Li Che, Xiaoyong Xu, Yujie Mu, Yongman Wu
Summary: Radar-based human motion recognition has gained much attention, but current methods can only recognize single actions and ignore the legality of continuous actions. This paper proposes a method using micro-Doppler features and Transformer for continuous action recognition, based on machine translation tasks and natural language processing. Additionally, an action state transition diagram is designed to ensure the legality of continuous actions. Experimental results show that the proposed method achieves accurate recognition for single actions and effective segmentation and recognition for continuous actions.
JOURNAL OF SENSORS
(2023)
Article
Engineering, Mechanical
Guillaume Girier, Mathieu Desroches, Serafim Rodrigues
Summary: Neuronal excitability is characterized by key markers of dynamics and allows for classification into different voltage response groups. Two main types of excitability and their mathematical models can be distinguished through bifurcation scenarios. Integrator and resonator neurons demonstrate differences in subthreshold oscillations. Switches between neural categories can be observed experimentally and modeled by bifurcation structure changes. This work proposes a novel scenario for switching between integrator and resonator neurons based on multiple-timescale dynamics, utilizing a specific time-dependent slowly varying current.
NONLINEAR DYNAMICS
(2023)
Article
Engineering, Electrical & Electronic
Qiang Wang, Gan Sun, Jiahua Dong, Qianqian Wang, Zhengming Ding
Summary: This paper proposes a lifelong multi-view subspace learning framework for continuous human action recognition, which utilizes complementary information among different views and achieves superior performance on new action recognition tasks.
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY
(2022)
Article
Biochemistry & Molecular Biology
Viktor Drgan, Benjamin Bajzelj
Summary: This study evaluated the abilities of different supervised self-organizing algorithms in classifying the hepatotoxic potential of drugs. Two modifications of standard algorithms were proposed and optimized models were evaluated, with the new algorithms showing better cluster formation compared to the traditional algorithm. The X-Y fused neural network demonstrated a high ability in forming well-separated clusters.
INTERNATIONAL JOURNAL OF MOLECULAR SCIENCES
(2021)
Article
Engineering, Chemical
Xing Wu, Hanlu Jin, Xueming Ye, Jianjia Wang, Zuosheng Lei, Ying Liu, Jie Wang, Yike Guo
Summary: The proposed MCRNN framework for reliable CCS quality prediction outperforms conventional methods by transforming input at different scales and frequencies, capturing both long-term trends and short-term changes in time series. Additionally, the framework generates different category distributions using the RUS method to mitigate the impact of skewed data distribution.
Article
Engineering, Electrical & Electronic
Lei Chen, Jiwen Lu, Zhanjie Song, Jie Zhou
Summary: In this paper, a recurrent semantic preserving generation method is proposed for action prediction, which achieves competitive performance in experimental results.
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY
(2021)
Review
Computer Science, Artificial Intelligence
AbdulRahman Baraka, Mohd Halim Mohd Noor
Summary: Temporal Action Localization is an important task in computer vision, involving video understanding, summarization, and analysis. With the high cost of temporal boundaries annotations, there is a growing interest in weakly-supervised Temporal Action Localization. This survey reviews the concepts, strategies, and techniques related to WTAL and presents state-of-the-art frameworks.
NEURAL COMPUTING & APPLICATIONS
(2022)
Review
Computer Science, Theory & Methods
Harshala Gammulle, David Ahmedt-Aristizabal, Simon Denman, Lachlan Tychsen-Smith, Lars Petersson, Clinton Fookes
Summary: In this paper, a comprehensive review of prediction models and action segmentation methods in video stream analysis is provided. The feature extraction and learning strategies used in state-of-the-art methods are thoroughly analyzed and compared. The impact of object detection and tracking techniques on human action segmentation is also discussed, as well as the limitations and key research directions for improving interpretability, generalization, optimization, and deployment.
ACM COMPUTING SURVEYS
(2023)
Article
Computer Science, Artificial Intelligence
Jiasen Wang, Jun Wang
Summary: This article presents a neurodynamic approach to nonlinear programming using a two-timescale multilayer recurrent neural network. The network has faster dynamics in the hidden layer(s) compared to the output layer. The article provides conditions for the convergence of the network to local optima of nonlinear programming problems and validates the approach through simulation results.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2022)
Article
Computer Science, Artificial Intelligence
Zicong Xia, Yang Liu, Jiasen Wang, Jun Wang
Summary: In this paper, the authors proposed two-timescale neurodynamic optimization approaches to distributed min-max optimization problems. They introduced four multilayer recurrent neural networks for solving different types of nonlinear convex-concave minimax problems subject to linear equality and nonlinear inequality constraints. The authors derived sufficient conditions to ensure the stability and optimality of the neural networks. They demonstrated the effectiveness and efficiency of the proposed neural networks in two specific paradigms for seeking Nash equilibrium in a zero-sum game and performing distributed constrained nonlinear optimization.
Article
Quantum Science & Technology
Ebrahim Ghasemian, Abolhassan Razminia, Habib Rostami
Summary: This paper proposes a realistic model for implementing neural networks on photonic quantum computers. A quantum circuit is designed using the continuous-variable architecture and encodes information in the spectral amplitude functions of single photons. The model is able to reproduce classical neural network models while maintaining quantum phenomena.
QUANTUM INFORMATION PROCESSING
(2023)
Article
Computer Science, Artificial Intelligence
Jie Fu, Junyu Gao, Changsheng Xu
Summary: This paper proposes a novel framework CRRC-Net for weakly-supervised temporal action localization. The framework addresses the issues of large intra-action variation and noisy classification learning through a co-supervised feature learning module and a probabilistic pseudo label mining module. Experimental results demonstrate the favorable performance of the method compared to state-of-the-art approaches.
IEEE TRANSACTIONS ON IMAGE PROCESSING
(2022)
Article
Computer Science, Artificial Intelligence
G. M. Mashrur E. Elahi, Yee-Hong Yang
Summary: Nowadays, deep learning methods have made significant advancements in human action recognition. However, processing the entire video sequence to recognize an action is unnecessary due to the similarity of many frames. To address this issue, keyframe-based methods have been proposed. Nevertheless, current methods still process all frames and average the results, leading to inaccurate recognition. To overcome this, a new online temporal classification model and action inference graph are proposed, enabling early recognition and reducing computation. The experimental results demonstrate the effectiveness of the proposed model in achieving state-of-the-art results without using full video sequences.
PATTERN RECOGNITION
(2022)
Article
Computer Science, Artificial Intelligence
Xiangtao Zheng, Tengfei Gong, Xiaoqiang Lu, Xuelong Li
Summary: This paper proposes a Multiple Spatial Clues Network (MSCNet) to represent spatial clues with image-level action labels, eliminating the need for labor-intensive annotations and additional supervision. The proposed MSCNet utilizes a spatial attention module to generate attention regions and detects spatial clues with minimal supervision. Experimental results demonstrate the effectiveness of the proposed MSCNet.
Article
Computer Science, Information Systems
Shaoyong Zhang, Na Li, Chenchen Qiu, Zhibin Yu, Haiyong Zheng, Bing Zheng
MULTIMEDIA TOOLS AND APPLICATIONS
(2020)
Article
Computer Science, Software Engineering
Haoxu Zhang, Chenchen Qiu, Chao Wang, Bin Wei, Zhibin Yu, Haiyong Zheng, Juan Li
Summary: This paper introduces a new method for generating 3D objects based on generative adversarial networks (GANs), utilizing multiple generators and discriminators to enhance learning complex distributions. The model employs spectral normalization technology to ensure stable training and generate high-quality and realistic 3D objects. Additionally, the system is capable of recovering incomplete 3D objects and outperforms baseline models in object quality.
CONCURRENCY AND COMPUTATION-PRACTICE & EXPERIENCE
(2021)
Article
Computer Science, Information Systems
Hao Ding, Bin Wei, Zhaorui Gu, Zhibin Yu, Haiyong Zheng, Bing Zheng, Juan Li
MULTIMEDIA TOOLS AND APPLICATIONS
(2020)
Article
Computer Science, Artificial Intelligence
Chao Wang, Wenjie Niu, Yufeng Jiang, Haiyong Zheng, Zhibin Yu, Zhaorui Gu, Bing Zheng
INTERNATIONAL JOURNAL OF COMPUTER VISION
(2020)
Article
Computer Science, Information Systems
Yan Zhao, Ziqiang Zheng, Chao Wang, Zhaorui Gu, Min Fu, Zhibin Yu, Haiyong Zheng, Nan Wang, Bing Zheng
MULTIMEDIA TOOLS AND APPLICATIONS
(2020)
Article
Chemistry, Analytical
Qingyun Li, Zhibin Yu, Yubo Wang, Haiyong Zheng
Article
Computer Science, Artificial Intelligence
Yubo Wang, Zhibin Yu, Tatinati Sivanagaraja, Kalyana C. Veluvolu
APPLIED SOFT COMPUTING
(2020)
Article
Computer Science, Hardware & Architecture
Ziqiang Zheng, Hongzhi Liu, Fan Yang, Xingyu Zheng, Zhibin Yu, Shaoda Zhang
Summary: This study introduces an innovative framework for photo-to-caricature translation, using a representation-guided scheme to mimic the caricature style, and introducing a feature-pyramid adversarial network to improve image synthesis quality. Experimental results demonstrate the excellent imitation capabilities of the proposed method across various caricature datasets.
COMPUTERS & ELECTRICAL ENGINEERING
(2021)
Article
Chemistry, Analytical
Qi Zhao, Zhichao Xin, Zhibin Yu, Bing Zheng
Summary: Estimation of underwater depth maps is crucial in underwater vision research, presenting challenges such as lack of paired data and dynamic underwater environments. Researchers have developed a novel framework combining image translation and depth map estimation techniques, utilizing a coarse-to-fine network for precise depth map estimation. The method efficiently addresses the issues in underwater image synthesis and depth map estimation, providing diverse underwater images and accurate depth map estimation results.
Article
Computer Science, Artificial Intelligence
Ziqiang Zheng, Zhibin Yu, Yang Wu, Haiyong Zheng, Bing Zheng, Minho Lee
Summary: This paper introduces a method to address the imbalanced learning problem through cross-species image-to-image translation, and proposes a novel, simple, and effective structure of Multi-Branch Discriminator (MBD) based on Generative Adversarial Networks (GANs). The effectiveness of the MBD is demonstrated through both theoretical analysis and empirical evaluation, showing remarkable performance in various cross-species image translation tasks.
Article
Computer Science, Information Systems
Ziqiang Zheng, Zhibin Yu, Haiyong Zheng, Yang Yang, Heng Tao Shen
Summary: The paper proposes an effective multi-adversarial framework based on part-global learning for one-shot cross-domain image-to-image translation. Extensive experiments show that the proposed approach achieves impressive results on imbalanced image domains and outperforms existing methods in one-shot image-to-image translation.
IEEE TRANSACTIONS ON MULTIMEDIA
(2022)
Article
Computer Science, Information Systems
Ruyue Han, Yang Guan, Zhibin Yu, Peng Liu, Haiyong Zheng
Article
Computer Science, Information Systems
Xinliang Zhang, Huimin Zeng, Xiang Liu, Zhibin Yu, Haiyong Zheng, Bing Zheng
Article
Computer Science, Information Systems
Huimin Zeng, Xinliang Zhang, Zhibin Yu, Yubo Wang
Article
Computer Science, Information Systems
Zhensheng Shi, Liangjie Cao, Cheng Guan, Haiyong Zheng, Zhaorui Gu, Zhibin Yu, Bing Zheng
Article
Computer Science, Artificial Intelligence
Hamdan Abdellatef, Lina J. Karam
Summary: This paper proposes performing the learning and inference processes in the compressed domain to reduce computational complexity and improve speed of neural networks. Experimental results show that modified ResNet-50 in the compressed domain is 70% faster than traditional spatial-based ResNet-50 while maintaining similar accuracy. Additionally, a preprocessing step with partial encoding is suggested to improve resilience to distortions caused by low-quality encoded images. Training a network with highly compressed data can achieve good classification accuracy with significantly reduced storage requirements.
Article
Computer Science, Artificial Intelligence
Victor R. Barradas, Yasuharu Koike, Nicolas Schweighofer
Summary: Inverse models are essential for human motor learning as they map desired actions to motor commands. The shape of the error surface and the distribution of targets in a task play a crucial role in determining the speed of learning.
Article
Computer Science, Artificial Intelligence
Ting Zhou, Hanshu Yan, Jingfeng Zhang, Lei Liu, Bo Han
Summary: We propose a defense strategy that reduces the success rate of data poisoning attacks in downstream tasks by pre-training a robust foundation model.
Article
Computer Science, Artificial Intelligence
Hao Sun, Li Shen, Qihuang Zhong, Liang Ding, Shixiang Chen, Jingwei Sun, Jing Li, Guangzhong Sun, Dacheng Tao
Summary: In this paper, the convergence rate of AdaSAM in the stochastic non-convex setting is analyzed. Theoretical proof shows that AdaSAM has a linear speedup property and decouples the stochastic gradient steps with the adaptive learning rate and perturbed gradient. Experimental results demonstrate that AdaSAM outperforms other optimizers in terms of performance.
Article
Computer Science, Artificial Intelligence
Juntong Yun, Du Jiang, Li Huang, Bo Tao, Shangchun Liao, Ying Liu, Xin Liu, Gongfa Li, Disi Chen, Baojia Chen
Summary: In this study, a dual manipulator grasping detection model based on the Markov decision process is proposed. By parameterizing the grasping detection model of dual manipulators using a cross entropy convolutional neural network and a full convolutional neural network, stable grasping of complex multiple objects is achieved. Robot grasping experiments were conducted to verify the feasibility and superiority of this method.
Article
Computer Science, Artificial Intelligence
Miaohui Zhang, Kaifang Li, Jianxin Ma, Xile Wang
Summary: This paper proposes an unsupervised person re-identification (Re-ID) method that uses two asymmetric networks to generate pseudo-labels for each other by clustering and updates and optimizes the pseudo-labels through alternate training. It also designs similarity compensation and similarity suppression based on the camera ID of pedestrian images to optimize the similarity measure. Extensive experiments show that the proposed method achieves superior performance compared to state-of-the-art unsupervised person re-identification methods.
Article
Computer Science, Artificial Intelligence
Florian Bacho, Dominique Chu
Summary: This paper proposes a new approach called the Forward Direct Feedback Alignment algorithm for supervised learning in deep neural networks. By combining activity-perturbed forward gradients, direct feedback alignment, and momentum, this method achieves better performance and convergence speed compared to other local alternatives to backpropagation.
Article
Computer Science, Artificial Intelligence
Xiaojian Ding, Yi Li, Shilin Chen
Summary: This research paper addresses the limitations of recursive feature elimination (RFE) and its variants in high-dimensional feature selection tasks. The proposed algorithms, which introduce a novel feature ranking criterion and an optimal feature subset evaluation algorithm, outperform current state-of-the-art methods.
Article
Computer Science, Artificial Intelligence
Naoko Koide-Majima, Shinji Nishimoto, Kei Majima
Summary: Visual images observed by humans can be reconstructed from brain activity, and the visualization of arbitrary natural images from mental imagery has been achieved through an improved method. This study provides a unique tool for directly investigating the subjective contents of the brain.
Article
Computer Science, Artificial Intelligence
Huanjie Tao, Qianyue Duan
Summary: In this paper, a hierarchical attention network with progressive feature fusion is proposed for facial expression recognition (FER), addressing the challenges posed by pose variation, occlusions, and illumination variation. The model achieves enhanced performance by aggregating diverse features and progressively enhancing discriminative features.
Article
Computer Science, Artificial Intelligence
Zhenyi Wang, Pengfei Yang, Linwei Hu, Bowen Zhang, Chengmin Lin, Wenkai Lv, Quan Wang
Summary: In the face of the complex landscape of deep learning, we propose a novel subgraph-level performance prediction method called SLAPP, which combines graph and operator features through an innovative graph neural network called EAGAT, providing accurate performance predictions. In addition, we introduce a mixed loss design with dynamic weight adjustment to improve predictive accuracy.
Article
Computer Science, Artificial Intelligence
Yiyang Yin, Shuangling Luo, Jun Zhou, Liang Kang, Calvin Yu-Chian Chen
Summary: Medical image segmentation is crucial for modern healthcare systems, especially in reducing surgical risks and planning treatments. Transanal total mesorectal excision (TaTME) has become an important method for treating colon and rectum cancers. Real-time instance segmentation during TaTME surgeries can assist surgeons in minimizing risks. However, the dynamic variations in TaTME images pose challenges for accurate instance segmentation.
Article
Computer Science, Artificial Intelligence
Teng Cheng, Lei Sun, Junning Zhang, Jinling Wang, Zhanyang Wei
Summary: This study proposes a scheme that combines the start-stop point signal features for wideband multi-signal detection, called Fast Spectrum-Size Self-Training network (FSSNet). By utilizing start-stop points to build the signal model, this method successfully solves the difficulty of existing deep learning methods in detecting discontinuous signals and achieves satisfactory detection speed.
Article
Computer Science, Artificial Intelligence
Wenming Wu, Xiaoke Ma, Quan Wang, Maoguo Gong, Quanxue Gao
Summary: The layer-specific modules in multi-layer networks are critical for understanding the structure and function of the system. However, existing methods fail to accurately characterize and balance the connectivity and specificity of these modules. To address this issue, a joint learning graph clustering algorithm (DRDF) is proposed, which learns the deep representation and discriminative features of the multi-layer network, and balances the connectivity and specificity of the layer-specific modules through joint learning.
Article
Computer Science, Artificial Intelligence
Guanghui Yue, Guibin Zhuo, Weiqing Yan, Tianwei Zhou, Chang Tang, Peng Yang, Tianfu Wang
Summary: This paper proposes a novel boundary uncertainty aware network (BUNet) for precise and robust colorectal polyp segmentation. BUNet utilizes a pyramid vision transformer encoder to learn multi-scale features and incorporates a boundary exploration module (BEM) and a boundary uncertainty aware module (BUM) to handle boundary areas. Experimental results demonstrate that BUNet outperforms other methods in terms of performance and generalization ability.