MSU Deinterlacer Benchmark 2020

# 1 Benchmark of deinterlacing methods

What’s new

Key features of the Benchmark

To submit deinterlacing method, please, follow 3 simple steps in the the Deinterlacer Submission section

We appreciate new ideas. Please, write us an e-mail to deinterlacer-benchmark@videoprocessing.ml

Leaderboard

The table below shows a comparison of deinterlacers by PSNR, SSIM metrics and by speed.

Click on the labels to sort the table

Rank Name PSNR SSIM FPS on CPU
1.0 MSU Deinterlacer 40.708 0.983 1.3
2.5 VapourSynth TDeintMod 39.916 0.977 50.29
3.0 NNEDI 39.625 0.978 1.91
4.0 Bob-Weave Deinterlacer 39.679 0.976 46.45
4.5 VapourSynth EEDI3 39.373 0.977 51.9
6.0 Real-Time Deep Deinterlacer 39.203 0.976 0.27
7.5 Bob 38.499 0.975 52.83
8.5 Weston 3-Field Deinterlacer 38.680 0.969 36.75
9.0 Kernel Deinterlacer (optimal parameters) 38.103 0.970 37.91
9.0 Elemental Live Low Latency Interpolation 38.056 0.972 Hardware Real-Time
11.0 YADIF 37.742 0.965 48.96
12.0 Elemental Live Motion Adaptive Interpolation 37.063 0.964 Hardware Real-Time
13.5 Kernel Deinterlacer 36.731 0.960 37.85
14.5 Studio Coast Pty vMix 36.990 0.950 Hardware Real-Time
14.5 Adobe Premiere Pro Built-In 36.092 0.958 43.82
16.0 Motion and Area Pixel Deinterlacer 35.415 0.950 2.15
16.5 Muksun Deinterlacer 35.444 0.949 1.95
18.0 PAL Interpolation 33.111 0.913 2.85
19.5 Elemental Live Motion Adaptive Blend 29.744 0.868 Hardware Real-Time
20.0 ASVZZZ Deinterlacer 27.499 0.902 1.9
20.5 Motion Compensation Deinterlacer 27.899 0.804 1.45

Full FrameRate Leaderboard

Rank Name PSNR SSIM FPS on CPU
1.0 MSU Deinterlacer 40.917 0.983 1.3
2.0 VapourSynth TDeintMod 40.071 0.978 50.29
3.5 VapourSynth EEDI3 39.547 0.978 51.9
4.0 Bob-Weave Deinterlacer 39.775 0.976 46.45
4.5 Real-Time Deep Deinterlacer 39.450 0.977 0.27
6.5 Bob 38.645 0.975 52.83
7.0 Weston 3-Field Deinterlacer 38.726 0.969 36.75
7.5 Elemental Live Low Latency Interpolation 37.908 0.972 Hardware Real-Time
9.0 YADIF 37.860 0.966 48.96
10.0 Elemental Live Motion Adaptive Interpolation 36.953 0.964 Hardware Real-Time
11.0 Studio Coast Pty vMix 32.942 0.931 Hardware Real-Time
12.0 Elemental Live Motion Adaptive Blend 29.747 0.868 Hardware Real-Time

Visualization

In this section you can see a frame, a crop from this frame, and also MSU VQMT PSNR Visualization of this crop.

Drag a red rectangle in the area which you want to crop, by default it is in the area with the worst PSNR

The area to compare on: Deinterlacer 1: Deinterlacer 2:

GT

In this row you can see VQMT PSNR Visualization

NNEDI

MSU Deinterlacer

VS TDeintMod

Charts

Highlight the plot region where you want to zoom in

Metric:

FPS is calculated on Intel-Core i7 10700K CPU

Metric:

The following plot shows difference between every method and Bob, because Bob is considered as the least complicated deinterlacing method

Metric:

Sequence №:

Sequence №: Metric:

Cross-PSNR

In this section you can see PSNR between the output of chosen deinterlacer and the others

Deinterlacer:

PSNR
ASVZZZ 20.54547840460791
Bob inf
Bob-Weave 42.07810541012702
Deep 41.164435818653764
EL LLI 43.87789597826898
EL MAI 38.01828007878827
Kernel 42.082198877630226
MAP 36.81130955719863
MSU 39.86926525444535
Muksun 37.93674757748442
NNEDI 43.222444723870076
PAL Interpolation 34.30022078877691
VMix 30.93623525121193
Weston 3-Field 42.526302640037
YADIF 39.61242962714242

Evaluation methodology

Coming soon!

Cite us

Coming soon!

Deinterlacer Submission

There are 3 easy steps to submit:

  1. Download the interlaced video here.
    We have more available formats, if YUV is not suitable. Just click on this text
    There are 5 available options:
      a. Download frames of the video sequence in .png format here
      b. Download .yuv video file generated from frames via

      ffmpeg -i %04d.png -c:v rawvideo -pix_fmt yuv444p sequences.yuv

      here
      c. Download lossless encoded .mkv video generated from frames via

      ffmpeg -i %04d.png -c:v libx264 -preset ultrafast -crf 0 -pix_fmt yuv444p lossless.mkv

      here
      d. Download .rgb video file generated from frames via

      ffmpeg -i %04d.png -c:v rawvideo -pix_fmt rgb24 sequences.rgb

      here
      e. Download lossless encoded .avi video generated from frames via

      ffmpeg -i %04d.png -c:v libx264rgb -preset ultrafast -crf 0 lossless.avi

      here

  2. Deinterlace downloaded video
    The details, which may help you
      TFF interlacing was used to get interlaced sequence from GT
      The video consists of 40 videos, each separated by 5 black frames. Black frames are ignored while measuring

  3. Send us an email to deinterlacer-benchmark@videoprocessing.ml with the following information:
      A. Name of the deinterlacing method that will be specified in our benchmark

      B. Link to the cloud drive (Google Drive, OneDrive, Dropbox, etc.), containing deinterlaced video.

      C. (Optional) Any additional information about the method
      Click here to see what may be included in additional information
        Technical information about deinterlaced video (e.g. colorspace, file-type, codec)
        The name of the theoretical method used
        Full name of the deinterlacing method or product
        The version that was used
        The parameter set that was used
        Any other additional information
        A link to the code of your deinterlacing method, if it is open-source
        A link to the paper about your deinterlacing method
        A link to the documentation of your deinterlacing method. For example, this is suitable for deinterlacing methods that are implemented as a part of a video processing framework.
        A link to the page, where users can purchase or download your product (for example, VirtualDub Plugin)

      D. (Optional) If you would like us to tune the parameters of your deinterlacing method, you should give us an ability to launch it. You can do it by sending us a code or an executable file, providing us a free test version of your product, or any other possible way, that is convenient for you

Contacts

For questions and propositions, please contact us: deinterlacer-benchmark@videoprocessing.ml

05 Nov 2020
See Also
MSU 3D-video Quality Analysis. Report 10
MSU SBD Benchmark 2020
MSU SBD Benchmark 2020
MSU Deinterlacer Benchmark 2020
Detection of stereo window violation
How to find objects that are present only in one view?
Depth continuity estimation in S3D video
How smooth is the depth transition between scenes?
Site structure