.png){kind=icon}
返回
Super-resolution reconstructed video coding scheme based on inter-frame information
.png){kind=icon}
PRE
AI
DOI：10.1007/s11042-023-17441-z.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
In recent years, the explosive growth of video data has posed a challenge to the performance of traditional video coding frameworks. The industry is thus faced with the pressing issue of how to transmit high-quality videos under limited bandwidth conditions. To address this challenge, a super-resolution reconstruction video coding scheme has been proposed, which combines traditional coding frameworks with deep learning-based super-resolution technology. By reducing the data coding volume through pre-coding downsampling and reconstructing videos through post-decoding upsampling, this scheme shows great potential in solving the aforementioned problems. However, previous super-resolution reconstruction video coding schemes have failed to effectively utilize the inter-frame correlation of video sequences, which limits the coding efficiency of the scheme. To overcome this limitation, this paper proposes an upsampling reconstruction network based on inter-frame information exploration. Experimental results show that, compared with the HEVC standard, the proposed scheme achieves a reduction of 11.05%, 11.3%, and 8.84% in the BD-BR index under the All-Intra, Low Delay P, and Random Access coding configurations, respectively, demonstrating higher coding efficiency than the previous human-designed scheme.
Keyword:
Video coding scheme
Super-resolution
Inter-frame correlation
HEVC
期刊
IF:
3
论文数:
1.9W
被引数:
3.2W
机构
引用论文
.png){kind=icon}
分享收藏分享收藏Simultaneous bilateral thoracoscopic lobectomy for synchronous bilateral multiple primary lung cancer—single center experience同期双侧胸腔镜肺叶切除术治疗同期双侧多原发肺癌—单中心经验
OAAI
OAAI分享收藏
查看更多内容