Article
Computer Science, Information Systems
Kai Lin, Chuanmin Jia, Xinfeng Zhang, Shanshe Wang, Siwei Ma, Wen Gao
Summary: This article introduces a new deep learning network structure for in-loop filtering in video coding. The proposed method utilizes the correlation between different color components and achieves improvements in both performance and efficiency through techniques such as adaptive granularity optimization and parallel inference pipeline.
ACM TRANSACTIONS ON MULTIMEDIA COMPUTING COMMUNICATIONS AND APPLICATIONS
(2022)
Article
Engineering, Electrical & Electronic
Mohammad Jubran, Alhabib Abbas, Yiannis Andreopoulos
Summary: This study proposes a stitching method that extends the frame referencing interval of inter-frame video coding to the entire length of video sequences. The method selects reference frames using compact feature descriptors and a similarity scoring mechanism, achieving significant rate gains with manageable complexity and memory requirements.
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY
(2022)
Article
Engineering, Electrical & Electronic
Jaehyun Bae, Sang-hyo Park
Summary: Video super-resolution (VSR) has been enhanced using deep learning architectures and clean image datasets, but the compression of real-world videos presents a challenge due to variation in frame quality. By exploiting a simple coding prior that identifies outstanding frames, we propose a method called YOLOF that enhances existing VSR models without compromising their original architectures. Extensive evaluations show that YOLOF substantially improves the performance of VSR models by inputting the highest quality frames near the reference frame with a given distance.
JOURNAL OF ELECTRONIC IMAGING
(2023)
Article
Computer Science, Artificial Intelligence
Zhijie Huang, Jun Sun, Xiaopeng Guo, Mingyu Shang
Summary: This paper proposes an adaptive deep reinforcement learning-based in-loop filter for versatile video coding. By developing a network set and training an agent network, the model can adapt to various video contents, enhancing robustness through a two-stage training scheme. Achieving better performance with low computation complexity compared to other deep learning-based methods, the proposed approach shows significant improvements in coding efficiency and subjective quality in video coding tasks.
IEEE TRANSACTIONS ON IMAGE PROCESSING
(2021)
Article
Engineering, Electrical & Electronic
Gang He, Li Xu, Jie Lei, Weiying Xie, Yunsong Li, Yibo Fan, Jinjia Zhou
Summary: This paper introduces a deep neural network for scalable high efficiency video coding, which improves visual quality and coding efficiency through interlayer restoration. By utilizing reconstructed frames from different layers, the network generates interlayers with higher quality, enhancing the coding efficiency.
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY
(2022)
Article
Computer Science, Artificial Intelligence
Xuewei Meng, Chuanmin Jia, Xinfeng Zhang, Shanshe Wang, Siwei Ma
Summary: This paper proposes a deformable Wiener Filter (DWF) that combines local and non-local characteristics for noise reduction in video coding. By training the filter coefficients based on the Wiener Filter theory, the DWF achieves superior performance compared to existing methods. The simulation results show that the proposed approach achieves an average bit-rate savings of 1.16% to 2.67% compared to the VTM-11.0.
IEEE TRANSACTIONS ON IMAGE PROCESSING
(2022)
Article
Computer Science, Artificial Intelligence
Weiheng Sun, Xiaohai He, Chao Ren, Shuhua Xiong, Honggang Chen
Summary: This article presents a novel Constant bit rate Video quality enhancement Network combined with Coding priors (CVCN) that aims to improve video quality in constant bit rate (CBR) coding mode. The CVCN model can be inserted into the High Efficiency Video Coding (HEVC) codec as a CNN-based in-loop filter (LF) module or a post-processing module. It utilizes CU-wise QP adaptability and CU-partition priors to handle the diversity of CBR videos. Experimental results demonstrate the superior performance of CVCN in enhancing CBR videos.
KNOWLEDGE-BASED SYSTEMS
(2022)
Article
Biology
Betul Ay, Cihan Turker, Elif Emre, Kevser Ay, Galip Aydin
Summary: This paper introduces the characteristics and harm of nasal polyps and proposes a reliable rhinology assistance system for recognizing them. The authors design a new dataset including 80 participants and conduct experiments using machine learning and deep learning algorithms. They find that deep learning algorithms achieve high accuracy in identifying nasal polyps. The research results are significant for supporting clinical decision systems.
COMPUTERS IN BIOLOGY AND MEDICINE
(2022)
Article
Engineering, Electrical & Electronic
Shiba Kuanar, Vassilis Athitsos, Dwarikanath Mahapatra, K. R. Rao
Summary: In order to achieve higher coding efficiency, the VVC standard introduces new components at the expense of increased computational complexity. However, these technologies often lead to visual artifacts and blurring. This paper proposes a deep-learning-based filtering model that utilizes feature correlation and multi-scale convolutional neural networks to enhance frame quality.
SIGNAL PROCESSING-IMAGE COMMUNICATION
(2022)
Article
Engineering, Electrical & Electronic
Ge Li, Jianjun Lei, Zhaoqing Pan, Bo Peng, Nam Ling
Summary: This paper proposes a multi-resolution prediction method for optimizing the coding efficiency of depth video, including selective encoding and deep up-sampling network for resolution recovery. Experimental results demonstrate that the proposed method achieves an average BD-rate saving of 10.84% compared to traditional methods.
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY
(2022)
Article
Engineering, Electrical & Electronic
Mingxuan Li, Wen Ji
Summary: A lightweight multiattention recursive residual CNN-based in-loop filter is proposed in this paper to handle encoded frames with various QP values, frame types (FTs), and temporal layers (TLs) via a single model. By learning multiscale features, adopting a recursive structure, and introducing auxiliary parameter fusion attention and long-short-term skip connection models, the proposed method achieves significant BD rate savings on standard test sequences, outperforming other state-of-the-art approaches.
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY
(2023)
Article
Computer Science, Artificial Intelligence
Dezhao Wang, Sifeng Xia, Wenhan Yang, Jiaying Liu
Summary: This paper addresses joint spatial-temporal modeling and side information injection for deep-learning based in-loop filter, improving the quality of reconstructed frames.
IEEE TRANSACTIONS ON IMAGE PROCESSING
(2021)
Article
Computer Science, Information Systems
Yong-Woon Kim, Yung-Cheol Byun, Dong Seog Han, Dalia Dominic, Sibu Cyriac
Summary: This paper proposes an Adaptive N-Frames Ensemble (AFE) approach for high-movement human segmentation in a video using an ensemble of multiple DL models. The AFE approach addresses the limitations of the existing methods and achieves good results in both low and high-movement human segmentation.
CMC-COMPUTERS MATERIALS & CONTINUA
(2022)
Article
Computer Science, Artificial Intelligence
Jinjin Xu, Wenli Du, Yaochu Jin, Wangli He, Ran Cheng
Summary: Federated learning is a privacy-preserving and secure machine learning approach. This paper proposes a federated trained ternary quantization algorithm and a ternary federated averaging protocol to reduce communication costs and improve performance.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2022)
Article
Computer Science, Information Systems
Jude Tchaye-Kondi, Yanlong Zhai, Jun Shen, Dong Lu, Liehuang Zhu
Summary: This article introduces SmartFilter, a new Edge-to-Cloud filtering solution for video analytics. It improves system performance, reduces server processing overhead, and maintains accuracy by filtering frames on the camera based on feedback from the server-side application.
IEEE INTERNET OF THINGS JOURNAL
(2022)
Article
Engineering, Electrical & Electronic
Ren Yang, Fabian Mentzer, Luc Van Gool, Radu Timofte
Summary: This paper introduces a Recurrent Learned Video Compression (RLVC) approach with Recurrent Auto-Encoder (RAE) and Recurrent Probability Model (RPM) to better utilize the temporal correlation among video frames, achieving state-of-the-art compression performance.
IEEE JOURNAL OF SELECTED TOPICS IN SIGNAL PROCESSING
(2021)
Article
Computer Science, Artificial Intelligence
Li Yang, Mai Xu, Yichen Guo, Xin Deng, Fangyuan Gao, Zhenyu Guan
Summary: This paper proposes a novel approach for modeling human attention on omnidirectional images (ODIs) by integrating hierarchical Bayesian inference and long short-term memory (LSTM) network. The approach predicts head trajectories and saliency on ODIs by capturing temporal correlations and modeling inter-subject uncertainty. Extensive experiments demonstrate the superior performance of the proposed approach compared to state-of-the-art methods.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2022)
Article
Computer Science, Artificial Intelligence
Zhenyu Guan, Junpeng Jing, Xin Deng, Mai Xu, Lai Jiang, Zhou Zhang, Yipeng Li
Summary: This paper proposes a novel multiple image hiding framework DeepMIH based on invertible neural network. An invertible hiding neural network (IHNN) is developed to model the image concealing and revealing as its forward and backward processes innovatively, making them fully coupled and reversible. In addition, an importance map (IM) module is designed to guide the current image hiding and enhance the invisibility. Experimental results show that DeepMIH significantly outperforms other state-of-the-art methods in terms of hiding invisibility, security and recovery accuracy.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2023)
Article
Engineering, Electrical & Electronic
Zhuojun Chen, Wenshang Lee, Qinhui Hong, Chongyan Gu, Zhenyu Guan, Lin Ding, Jiliang Zhang
Summary: The paper presents a hardware security mechanism called OFSR PUF, which consists of weak PUF cells and an obfuscation mechanism. It effectively reduces storage overhead, overcomes collapse response, and provides higher security.
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS II-EXPRESS BRIEFS
(2022)
Article
Computer Science, Information Systems
Feifei Liu, Yu Yan, Yu Sun, Jianwei Liu, Dawei Li, Zhenyu Guan
Summary: This paper proposes a PUF-based batch authentication and key agreement protocol to protect both meters and gateways in the smart grid and provide end-to-end authentication. The computation overhead is reduced significantly by offloading heavy operations from field devices to the server. Additionally, devolving batch authentication and access control to the gateway decreases downlink communication and signaling cost and surpasses most recent schemes.
SECURITY AND COMMUNICATION NETWORKS
(2022)
Article
Engineering, Electrical & Electronic
Ren Yang, Radu Timofte, Luc Van Gool
Summary: In recent years, there has been an increasing interest in end-to-end learned video compression. Previous works focused on compressing motion maps to exploit temporal redundancy. However, they did not fully utilize historical information in sequential reference frames. This paper proposes an Advanced Learned Video Compression (ALVC) approach with an in-loop frame prediction module, which effectively predicts the target frame from previously compressed frames. The experiments demonstrate the state-of-the-art performance of ALVC in learned video compression.
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY
(2023)
Article
Computer Science, Hardware & Architecture
Hua Deng, Zheng Qin, Qianhong Wu, Robert H. Deng, Zhenyu Guan, Yupeng Hu, Fangmin Li
Summary: Cloud computing is popular for data storage and sharing. Encryption is important for data security, but can hinder data sharing. This article proposes a hierarchical data sharing scheme that allows the data owner to selectively share encrypted data with users in a hierarchy, providing control over access.
IEEE TRANSACTIONS ON DEPENDABLE AND SECURE COMPUTING
(2023)
Article
Engineering, Electrical & Electronic
Wanyu Hou, Yu Sun, Dawei Li, Zhenyu Guan, Jianwei Liu
Summary: This paper introduces a 5G-V2G system that integrates the 5G network and the power grid to enable the remote transmission of reservation information. A scalable reservation authentication and key agreement protocol is proposed to support flexible access and ensure the confidentiality and privacy of critical reservation data. The protocol utilizes lightweight algorithms and physical unclonable function (PUF) for security, outperforming existing schemes in terms of computing overhead, transmission overhead, signaling overhead, and security.
IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY
(2023)
Article
Engineering, Electrical & Electronic
Xin Deng, Yufan Deng, Ren Yang, Wenzhe Yang, Radu Timofte, Mai Xu
Summary: In this paper, a novel network model called MASIC is proposed for stereo image compression. It achieves higher compression efficiency and quality through the introduction of a mask prediction module and mask conditional stereo entropy model.
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY
(2023)
Article
Automation & Control Systems
Hanzhou Wang, Dongyu Li, Zhenyu Guan, Yizhong Liu, Jianwei Liu
Summary: This paper introduces a framework to protect privacy in multi-agent systems, allowing the states to reach a consensus while keeping the initial states confidential. A privacy-preserving two-party relationship test protocol is proposed, which is then used to devise average consensus and rendezvous controllers for first- and second-order systems. Unlike previous research that relies on stochastic coupling weights, our approach overcomes the random chattering problem in control input, leading to improved convergence performance. Numerical verification is conducted to demonstrate the effectiveness of the proposed controllers.
IEEE CONTROL SYSTEMS LETTERS
(2023)
Article
Computer Science, Theory & Methods
Yizhong Liu, Xinxin Xing, Haosu Cheng, Dawei Li, Zhenyu Guan, Jianwei Liu, Qianhong Wu
Summary: This paper proposes a flexible sharding (FS) blockchain protocol that addresses the drawbacks of existing sharding blockchain schemes through the design of a cross-shard Byzantine fault tolerance protocol, multiple parallel CSBFT, a defense mechanism against cross-shard transaction censorship attacks, and a secure and decentralized shard reconfiguration method. The paper also provides a formal protocol design method and security proofs for each protocol. The evaluation shows that FS has lower complexity and achieves significant performance improvements.
IEEE TRANSACTIONS ON INFORMATION FORENSICS AND SECURITY
(2023)
Article
Engineering, Civil
Dawei Li, Ruonan Chen, Qinjun Wan, Zhenyu Guan, Shizhong Li, Min Xie, Jieyu Su, Jianwei Liu
Summary: The emergence of electric vehicles has led to the development of the Internet of Vehicles (IoV), but there are various security issues that need to be addressed. This paper proposes a blockchain-based intelligent and fair IoV charging service system that intelligently recommends charging piles for vehicles and ensures fairness between charging and payment through a payment channel protocol.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2023)
Article
Engineering, Electrical & Electronic
Xin Deng, Bihe Zhao, Zhenyu Guan, Mai Xu
Summary: Recently, fake face images generated by generative adversarial network (GAN) have become a major concern in social networks due to the security risks they pose. This study reveals that GAN generated fake images have stronger non-local self-similarity than real images, leading to the development of NAFID, a non-local attention based fake image detection network. Experimental results demonstrate the superiority of NAFID over state-of-the-art face forgery detection methods and the potential to improve the detection accuracy of other models.
IEEE SIGNAL PROCESSING LETTERS
(2023)
Proceedings Paper
Computer Science, Artificial Intelligence
Yannick Strumpler, Janis Postels, Ren Yang, Luc Van Gool, Federico Tombari
Summary: This study introduces a compression method based on Implicit Neural Representations (INRs), which significantly improves compression quality and outperforms traditional algorithms through meta-learned initializations and network structure improvements.
COMPUTER VISION, ECCV 2022, PT XXVI
(2022)
Proceedings Paper
Computer Science, Theory & Methods
Ren Yang, Radu Timofte, Meisong Zheng, Qunliang Xing, Minglang Qiao, Mai Xu, Lai Jiang, Huaida Liu, Ying Chen, Youcheng Ben, Xiao Zhou, Chen Fu, Pei Cheng, Gang Yu, Junyi Li, Renlong Wu, Zhilu Zhang, Wei Shang, Zhengyao Lv, Yunjin Chen, Mingcai Zhou, Dongwei Ren, Kai Zhang, Wangmeng Zuo, Pavel Ostyakov, Vyal Dmitry, Shakarim Soltanayev, Chervontsev Sergey, Zhussip Magauiya, Xueyi Zou, Youliang Yan, Pablo Navarrete Michelini, Yunhua Lu, Diankai Zhang, Shaoli Liu, Si Gao, Biao Wu, Chengjian Zheng, Xiaofeng Zhang, Kaidi Lu, Ning Wang, Thuong Nguyen Canh, Thong Bach, Qing Wang, Xiaopeng Sun, Haoyu Ma, Shijie Zhao, Junlin Li, Liangbin Xie, Shuwei Shi, Yujiu Yang, Xintao Wang, Jinjin Gu, Chao Dong, Xiaodi Shi, Chunmei Nian, Dong Jiang, Jucai Lin, Zhihuai Xie, Mao Ye, Dengyan Luo, Liuhan Peng, Shengjie Chen, Xin Liu, Qian Wang, Boyang Liang, Hang Dong, Yuhao Huang, Kai Chen, Xingbei Guo, Yujing Sun, Huilei Wu, Pengxu Wei, Yulin Huang, Junying Chen, Ik Hyun Lee, Sunder Ali Khowaja, Jiseok Yoon
Summary: This paper reviews the NTIRE 2022 Challenge on Super-Resolution and Quality Enhancement of Compressed Video, introducing the dataset, tracks, participating teams, and final results. The challenge evaluates the state-of-the-art techniques in super-resolution and quality enhancement of compressed video, providing relevant datasets and code resources.
2022 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION WORKSHOPS, CVPRW 2022
(2022)