From Avisynth wiki
Jump to: navigation, search
Author Asd-g
Version v1.0.0
Download VMAF-
Category Debug filter
License MIT



VMAF is a perceptual video quality assessment algorithm developed by Netflix. Refer to the FAQ page for frequently asked questions of VMAF.

This is a port of the VapourSynth plugin VMAF.

vmaf is used.


NOTE:The folder model must be located in the same folder as VMAF.dll.

Syntax and Parameters

VMAF (clip reference, clip distorted, int "model", string "log_path", int "log_fmt", bool "ssim", bool "ms_ssim", int "pool", bool "ci")

clip   =
clip   =
Clips to calculate VMAF score. Must be in YUV 8..16-bit planar format.
First input clip is the reference, second input clip is the distorted clip (e.g. processed, encoded, etc).

int  model = 0
Sets which model to use. Refer to the models page for more details.
  • 0: vmaf_v0.6.1.pkl
  • 1: vmaf_4k_v0.6.1.pkl
Default: 0.

string  log_path =
Sets the path of the log file.
Default: None.

int  log_fmt = 0
Sets the format of the log file.
  • 0: xml.
  • 1: json.
  • 2: csv.
Default: 0.

bool  ssim = false
Whether to also calculate SSIM score.
Default: False.

bool  ms_ssim = false
Whether to also calculate MS-SSIM score.
Default: False.

int  pool = 1
Sets the method to pool the per-frame scores.
  • 0: Mean.
  • 1: Harmonic mean.
  • 2: Min.
Default: 1.

bool  ci = false
Whether to enable confidence interval.
True: It uses vmaf_b_v0.6.3 for model=0 and vmaf_4k_rb_v0.6.2 for model=1.
Refer to the VMAF confidence interval page for more details.
Default: False.


Comparing a high quality clip vs an encoded clip:

VMAF(HQ_original, HQ_encoded, model=0, ssim=false, ms_ssim=false, pool=1, ci=false)


Version      Date            Changes
v1.0.0 2020/11/14 - Initial release - AviSynth+: self-registers as MT_SERIALIZED

External Links

  • GitHub - Source code repository.

Back to External Filters

Personal tools