Scholarlink

This paper proposes Industrial AV-JCoder, a low-latency joint audio-video compression framework for IIoT applications. It integrates frequency-domain motion estimation and Light-VAE encoding for video, cross-modal attention-based denoising for audio, and adaptive bitrate control for stable transmission. Experimental results show that the proposed method outperforms H.265 + Opus and AV1 + Lyra in compression quality, real-time performance, and network adaptability, making it well-suited for 5G edge computing, industrial monitoring, and AR/VR streaming.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Low-Latency Joint Audio-Video Compression Framework for Real-Time Transmission

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Jie Zhou,
  • RuiFeng Ding,
  • ShengMing Li,
  • ZiYang Song,
  • Shengyang Gao,
  • Zhuang Shao,
  • Lijun Lei,
  • Benjun Dou

摘要

This paper proposes Industrial AV-JCoder, a low-latency joint audio-video compression framework for IIoT applications. It integrates frequency-domain motion estimation and Light-VAE encoding for video, cross-modal attention-based denoising for audio, and adaptive bitrate control for stable transmission. Experimental results show that the proposed method outperforms H.265 + Opus and AV1 + Lyra in compression quality, real-time performance, and network adaptability, making it well-suited for 5G edge computing, industrial monitoring, and AR/VR streaming.