[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-6539":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":10,"language":11,"languages":10,"totalLinesOfCode":10,"stars":12,"forks":13,"watchers":14,"openIssues":15,"contributorsCount":16,"subscribersCount":16,"size":16,"stars1d":16,"stars7d":17,"stars30d":18,"stars90d":16,"forks30d":16,"starsTrendScore":19,"compositeScore":20,"rankGlobal":10,"rankLanguage":10,"license":21,"archived":22,"fork":22,"defaultBranch":23,"hasWiki":22,"hasPages":22,"topics":24,"createdAt":10,"pushedAt":10,"updatedAt":25,"readmeContent":26,"aiSummary":27,"trendingCount":16,"starSnapshotCount":16,"syncStatus":28,"lastSyncTime":29,"discoverSource":30},6539,"vmaf","Netflix\u002Fvmaf","Netflix","Perceptual video quality assessment based on multi-method fusion.","",null,"C",5367,820,487,66,0,4,29,1,69.64,"Other",false,"master",[],"2026-06-12 04:00:29","# VMAF - Video Multi-Method Assessment Fusion\n\n[![libvmaf](https:\u002F\u002Fgithub.com\u002FNetflix\u002Fvmaf\u002Factions\u002Fworkflows\u002Flibvmaf.yml\u002Fbadge.svg)](https:\u002F\u002Fgithub.com\u002FNetflix\u002Fvmaf\u002Factions\u002Fworkflows\u002Flibvmaf.yml)\n[![Windows](https:\u002F\u002Fgithub.com\u002FNetflix\u002Fvmaf\u002Factions\u002Fworkflows\u002Fwindows.yml\u002Fbadge.svg)](https:\u002F\u002Fgithub.com\u002FNetflix\u002Fvmaf\u002Factions\u002Fworkflows\u002Fwindows.yml)\n[![ffmpeg](https:\u002F\u002Fgithub.com\u002FNetflix\u002Fvmaf\u002Factions\u002Fworkflows\u002Fffmpeg.yml\u002Fbadge.svg)](https:\u002F\u002Fgithub.com\u002FNetflix\u002Fvmaf\u002Factions\u002Fworkflows\u002Fffmpeg.yml)\n[![Docker](https:\u002F\u002Fgithub.com\u002FNetflix\u002Fvmaf\u002Factions\u002Fworkflows\u002Fdocker.yml\u002Fbadge.svg)](https:\u002F\u002Fgithub.com\u002FNetflix\u002Fvmaf\u002Factions\u002Fworkflows\u002Fdocker.yml)\n\nVMAF is an [Emmy-winning](https:\u002F\u002Ftheemmys.tv\u002F) perceptual video quality assessment algorithm developed by Netflix. This software package includes a stand-alone C library `libvmaf` and its wrapping Python library. The Python library also provides a set of tools that allows a user to train and test a custom VMAF model.\n\nRead [this](https:\u002F\u002Fmedium.com\u002Fnetflix-techblog\u002Ftoward-a-practical-perceptual-video-quality-metric-653f208b9652) tech blog post for an overview, [this](https:\u002F\u002Fmedium.com\u002Fnetflix-techblog\u002Fvmaf-the-journey-continues-44b51ee9ed12) post for the tips of best practices, and [this](https:\u002F\u002Fnetflixtechblog.com\u002Ftoward-a-better-quality-metric-for-the-video-community-7ed94e752a30) post for our latest efforts on speed optimization, new API design and the introduction of a codec evaluation-friendly [NEG mode](resource\u002Fdoc\u002Fmodels.md#disabling-enhancement-gain-neg-mode).\n\nAlso included in `libvmaf` are implementations of several other metrics: PSNR, PSNR-HVS, SSIM, MS-SSIM and CIEDE2000.\n\n![vmaf logo](resource\u002Fimages\u002Fvmaf_logo.jpg)\n\n## News\n\n- (2023-12-07) We are releasing `libvmaf v3.0.0`. It contains several optimizations and bug fixes, and a full removal of the APIs which were deprecated in `v2.0.0`.\n- (2021-12-15) We have added to CAMBI the `full_ref` input parameter to allow running CAMBI as a full-reference metric, taking into account the banding that was already present on the source. Check out the [usage](resource\u002Fdoc\u002Fcambi.md) page.\n- (2021-12-1) We have added to CAMBI the `max_log_contrast` input parameter to allow to capture banding with higher contrasts than the default. We have also sped up CAMBI (e.g., around 4.5x for 4k). Check out the [usage](resource\u002Fdoc\u002Fcambi.md) page.\n- (2021-10-7) We are open-sourcing CAMBI (Contrast Aware Multiscale Banding Index) - Netflix's detector for banding (aka contouring) artifacts. Check out the [tech blog](https:\u002F\u002Fnetflixtechblog.medium.com\u002Fcambi-a-banding-artifact-detector-96777ae12fe2) for an overview and the [technical paper](resource\u002Fdoc\u002Fpapers\u002FCAMBI_PCS2021.pdf) published in PCS 2021 (note that the paper describes an initial version of CAMBI that no longer matches the code exactly, but it is still a good introduction). Also check out the [usage](resource\u002Fdoc\u002Fcambi.md) page.\n- (2020-12-7) Check out our [latest tech blog](https:\u002F\u002Fnetflixtechblog.com\u002Ftoward-a-better-quality-metric-for-the-video-community-7ed94e752a30) on speed optimization, new API design and the introduction of a codec evaluation-friendly NEG mode.\n- (2020-12-3) We are releasing `libvmaf v2.0.0`. It has a new fixed-point and x86 SIMD-optimized (AVX2, AVX-512) implementation that achieves 2x speed up compared to the previous floating-point version. It also has a [new API](libvmaf\u002FREADME.md) that is more flexible and extensible.\n- (2020-7-13) We have created a [memo](https:\u002F\u002Fdocs.google.com\u002Fdocument\u002Fd\u002F1dJczEhXO0MZjBSNyKmd3ARiCTdFVMNPBykH4_HMPoyY\u002Fedit?usp=sharing) to share our thoughts on VMAF's property in the presence of image enhancement operations, its impact on codec evaluation, and our solutions. Accordingly, we have added a new mode called [No Enhancement Gain (NEG)](resource\u002Fdoc\u002Fmodels.md#disabling-enhancement-gain-neg-mode).\n- (2020-2-27) We have changed VMAF's license from Apache 2.0 to [BSD+Patent](https:\u002F\u002Fopensource.org\u002Flicenses\u002FBSDplusPatent), a more permissive license compared to Apache that also includes an express patent grant.\n\n## Documentation\n\nThere is an [overview of the documentation](resource\u002Fdoc\u002Findex.md) with links to specific pages, covering FAQs, available models and features, software usage guides, and a list of resources.\n\n## Usage\n\nThe software package offers a number of ways to interact with the VMAF implementation.\n\n  - The command-line tool [`vmaf`](libvmaf\u002Ftools\u002FREADME.md) provides a complete algorithm implementation, such that one can easily deploy VMAF in a production environment. Additionally, the `vmaf` tool provides a number of auxillary features such as PSNR, SSIM and MS-SSIM.\n  - The [C library `libvmaf`](libvmaf\u002FREADME.md) provides an interface to incorporate VMAF into your code, and tools to integrate other feature extractors into the library.\n  - The [Python library](resource\u002Fdoc\u002Fpython.md) offers a full array of wrapper classes and scripts for software testing, VMAF model training and validation, dataset processing, data visualization, etc.\n  - VMAF is now included as a filter in FFmpeg, and can be configured using: `.\u002Fconfigure --enable-libvmaf`. Refer to the [Using VMAF with FFmpeg](resource\u002Fdoc\u002Fffmpeg.md) page.\n  - [VMAF Dockerfile](Dockerfile) generates a docker image from the [Python library](resource\u002Fdoc\u002Fpython.md). Refer to [this](resource\u002Fdoc\u002Fdocker.md) document for detailed usage.\n  - To build VMAF on Windows, follow [these](resource\u002Fdoc\u002Fwindows.md) instructions.\n  - AOM CTC: [AOM]((http:\u002F\u002Faomedia.org\u002F)) has specified vmaf to be the standard implementation metrics tool according to the AOM common test conditions (CTC). Refer to [this page](resource\u002Fdoc\u002Faom_ctc.md) for usage compliant with AOM CTC.\n\n## Contribution Guide\n\nRefer to the [contribution](CONTRIBUTING.md) page. Also refer to this [slide deck](https:\u002F\u002Fdocs.google.com\u002Fpresentation\u002Fd\u002F1Gr4-MvOXu9HUiH4nnqLGWupJYMeh6nl2MNz6Qy9153c\u002Fedit#slide=id.gc20398b4b7_0_132) for an overview contribution guide.\n","VMAF 是由 Netflix 开发的一种基于多方法融合的感知视频质量评估算法。其核心功能包括一个独立的 C 语言库 `libvmaf` 及其 Python 包装库，支持用户训练和测试自定义 VMAF 模型。此外，`libvmaf` 还实现了 PSNR、PSNR-HVS、SSIM、MS-SSIM 和 CIEDE2000 等多种质量指标。该工具适用于需要精确评估视频编码或传输过程中画质损失情况的场景，如视频流媒体服务、视频压缩算法研究等。通过提供一种更贴近人类视觉感知的质量评价方式，VMAF 帮助开发者优化视频处理流程中的参数设置，从而提升最终用户体验。",2,"2026-06-11 03:07:32","top_language"]