相关参考文献
注意:仅列出部分参考文献,下载原文获取全部文献信息。
Article
Computer Science, Information Systems
Zhenxiao Luo et al.
Summary: This paper proposes a neural-enhanced HD live video streaming framework called LiveSR, which aims to provide universal HD live video streaming for both bandwidth-constrained and bandwidth-rich devices. For bandwidth-constrained devices, LiveSR delivers low-quality video streams and enhances video quality at the device side using super-resolution (SR) techniques. To overcome the challenges of training the SR model with low cost and conducting real-time quality enhancement, a crowdsourced online training method and imitation learning-based decision-making algorithm are introduced. Experimental results show that LiveSR outperforms other baseline approaches, with 65.5% improvement in average QoE and 5.7% improvement in video quality (PSNR), achieving a frame rate as high as 30 frames per second.
IEEE TRANSACTIONS ON MULTIMEDIA
(2023)
Article
Computer Science, Software Engineering
Alexandra Covaci et al.
Summary: Omnidirectional videos are widely used in Virtual Reality applications. This article explores the impact of quality degradation on the sense of presence and investigates the enhancement of 360 degrees technology through multisensory stimuli. The results show that multisensory content significantly improves the immersive experience.
IEEE TRANSACTIONS ON VISUALIZATION AND COMPUTER GRAPHICS
(2023)
Proceedings Paper
Computer Science, Artificial Intelligence
Kaiming He et al.
Summary: This paper presents a self-supervised learning method for computer vision based on masked autoencoders. By masking a portion of the input image and reconstructing the missing pixels, large models can be trained efficiently and effectively. The approach achieves high generalization performance and outperforms supervised pretraining in transfer learning tasks.
2022 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2022)
(2022)
Article
Computer Science, Artificial Intelligence
Zhihao Wang et al.
Summary: This article provides a comprehensive survey on recent advances of image super-resolution using deep learning approaches, categorizing existing studies into supervised, unsupervised, and domain-specific SR techniques, as well as covering benchmark datasets and evaluation metrics. Future directions and open issues in the field are also highlighted for further research.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2021)
Article
Computer Science, Software Engineering
David Li et al.
Summary: In this paper, a new log-rectilinear transformation method is introduced to enable foveated streaming of 360 degrees videos suitable for VR headsets with built-in eye-tracking. The method heavily reduces visual artifacts and flickering issues compared to traditional log-polar subsampling, while also yielding an additional 10% reduction in bandwidth usage.
IEEE TRANSACTIONS ON VISUALIZATION AND COMPUTER GRAPHICS
(2021)
Article
Engineering, Electrical & Electronic
Abid Yaqoob et al.
Summary: The CFOV solution proposed in this paper combines two FoV prediction mechanisms to improve user QoE and bitrate distribution for 360-degree video streaming. Experimental findings demonstrate higher viewport match and improved user QoE compared to alternative approaches.
IEEE TRANSACTIONS ON BROADCASTING
(2021)
Article
Engineering, Electrical & Electronic
Jin Wan et al.
Summary: Ultra-high-definition display technology is widely used in broadcasting but faces a contradiction between high-resolution content and storage capacity. Super-Resolution techniques can alleviate this issue, with recent advancements in Deep Convolutional Neural Networks showing high-quality reconstruction performance but facing limitations due to large model parameters.
IEEE TRANSACTIONS ON BROADCASTING
(2021)
Proceedings Paper
Computer Science, Artificial Intelligence
Jiaming Liu et al.
Summary: This paper explores the use of deep neural networks to enhance video delivery quality, with a method of compressing models for better super-resolution results and a new approach to video coding. Extensive experiments show the method's superior video quality performance, demonstrating significant potential.
2021 IEEE/CVF INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV 2021)
(2021)
Proceedings Paper
Computer Science, Artificial Intelligence
Lovish Chopra et al.
Summary: Advancements in 360-degree video technologies have led to a popular research topic in streaming such videos, requiring optimized algorithms due to high bandwidth needs. Existing research faces challenges in predicting user viewports, but PARIMA improves this by combining user viewports and video content effectively.
PROCEEDINGS OF THE WORLD WIDE WEB CONFERENCE 2021 (WWW 2021)
(2021)
Proceedings Paper
Computer Science, Hardware & Architecture
Chenglei Wu et al.
Summary: In this research, a dual-queue streaming framework is proposed to address the limitations of traditional tile-based 360 degrees video streaming methods. A preference-aware DRL algorithm is designed to incentivize efficient learning of preference-dependent ABR decisions by the agent. Compared with state-of-the-art DRL baselines, the proposed method significantly improves streaming quality and demonstrates better performance and generalization under dynamic preferences.
PROCEEDINGS OF THE 31ST ACM WORKSHOP ON NETWORK AND OPERATING SYSTEMS SUPPORT FOR DIGITAL AUDIO AND VIDEO (NOSSDAV '21)
(2021)
Article
Computer Science, Theory & Methods
Royson Lee et al.
Summary: This article explores the state-of-the-art content delivery systems that utilize neural enhancement for achieving fast response time and high visual quality. It highlights the importance of neural enhancement in addressing technical challenges and analyzes existing systems and design decisions in efficiently overcoming these challenges.
ACM COMPUTING SURVEYS
(2021)
Proceedings Paper
Computer Science, Hardware & Architecture
Yanpeng Cao et al.
Summary: This paper proposes a real-time VSR system EGVSR, which achieves high-quality super-resolution video processing capability through a lightweight network structure and efficient upsampling methods. In experiments, EGVSR has made significant progress in computation density, performance acceleration, and visual quality.
2021 IEEE 32ND INTERNATIONAL CONFERENCE ON APPLICATION-SPECIFIC SYSTEMS, ARCHITECTURES AND PROCESSORS (ASAP 2021)
(2021)
Article
Computer Science, Information Systems
Xin Liu et al.
Summary: Super-resolution (SR) technology is crucial for mobile applications, with existing algorithms usually operating in the cloud due to high memory requirements, which limits real-time performance. However, the innovative SplitSR approach enables on-device super-resolution, achieving higher accuracy and faster inference speeds.
PROCEEDINGS OF THE ACM ON INTERACTIVE MOBILE WEARABLE AND UBIQUITOUS TECHNOLOGIES-IMWUT
(2021)
Article
Engineering, Electrical & Electronic
Zhuqing Jiang et al.
IEEE TRANSACTIONS ON BROADCASTING
(2020)
Proceedings Paper
Computer Science, Hardware & Architecture
Jiawen Chen et al.
NOSSDAV '20: PROCEEDINGS OF THE 2020 WORKSHOP ON NETWORK AND OPERATING SYSTEM SUPPORT FOR DIGITAL AUDIO AND VIDEO
(2020)
Proceedings Paper
Computer Science, Hardware & Architecture
Mallesham Dasari et al.
IEEE INFOCOM 2020 - IEEE CONFERENCE ON COMPUTER COMMUNICATIONS
(2020)
Article
Engineering, Electrical & Electronic
Yongwoo Kim et al.
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY
(2019)
Proceedings Paper
Computer Science, Theory & Methods
Pan Hu et al.
HOTMOBILE '19 - PROCEEDINGS OF THE 20TH INTERNATIONAL WORKSHOP ON MOBILE COMPUTING SYSTEMS AND APPLICATIONS
(2019)
Proceedings Paper
Computer Science, Artificial Intelligence
Petr Kellnhofer et al.
2019 IEEE/CVF INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV 2019)
(2019)
Proceedings Paper
Computer Science, Artificial Intelligence
Joris Heyse et al.
2019 26TH IEEE CONFERENCE ON VIRTUAL REALITY AND 3D USER INTERFACES (VR)
(2019)
Article
Computer Science, Software Engineering
Taehyun Rhee et al.
IEEE TRANSACTIONS ON VISUALIZATION AND COMPUTER GRAPHICS
(2017)
Proceedings Paper
Computer Science, Artificial Intelligence
Christian Ledig et al.
30TH IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2017)
(2017)