Netflix open-sources VMAF v1 video quality model, keeps v0 docs available
What happened
Netflix released a new v1 model for VMAF, its open-source perceptual video quality assessment tool, and is keeping the older v0 model's documentation available. VMAF is an Emmy-winning multi-method fusion video quality algorithm; its package includes the libvmaf C library and a Python wrapper, and integrates PSNR, SSIM, MS-SSIM and CIEDE2000 metrics. Netflix says it will keep improving the model based on user feedback. The main progress in this update is the v1 release; the old docs remain readable.
Written by AI from the coverage · updated 2 hours ago
Coverage
Follow the reports to see the story from different sides.
- Hacker News front pageNetflix vmaf: Perceptual video quality assessment based on multi-method fusion
Netflix 发布新一代 VMAF 感知视频质量评估模型 v1，旧版 v0 模型文档继续保留，官方表示将持续根据用户反馈改进模型。VMAF 是曾获艾美奖的多方法融合视频质量算法，软件包包含 C 库 libvmaf 与 Python 封装库，并集成 PSNR、SSIM、MS-SSIM、CIEDE2000 等指标。
Heat over time
Not enough continuous observations to draw a trend yet.