MSU Video Upscalers Benchmark 2022: Quality Enhancement
The most comprehensive comparison of video super resolution (VSR) algorithms by subjective quality
Our benchmark will help you to select the best upscaling approach for your needs. Check the “Methodology” section to learn the details.
Everyone is welcome to participate! Run your favorite super-resolution method on our compact test video and send us the result to see how well it performs. Check the “Submitting” section to learn the details.
What’s new
- November 9th, 2021: Alpha-version Release
Key features of the Benchmark
- Subjective comparison with over 4300 participants
- Visual comparison for all 40 algorithms on all 15 clips
- Find the upscaler of the best visual quality
- Check how upscalers behave in the most practical upscale use cases
- Find the method which produces the best results on various content types
- Special calculation of objective metrics robust to object borders displacement
- Easily spot upscaling artifacts with metric error maps visualizations
Introduction
Our benchmark presents the ranking of video upscalers using crowd-sourced subjective comparison. Over 4300 valid participants have selected the most visually appealing upscaling result in a number of pairwise comparisons.
For the evaluation of upscaling algorithms various metrics (objective measures of quality) also were used, including VMAF developed by Netflix. In addition FPS (frames per second) values were calculated (algorithms ran on NVIDIA Titan Xp).
Check the “Participants” section to learn the details about used algorithms.
4× Leaderboards
4× Visualizations
Clip:
Frame:
Error map:
Model 1:
Model 2:
Model 3:
Drag a red rectangle to the area that you want to see zoomed-in
Source
undef
undef
undef
4× Charts
2× Leaderboards
2× Visualizations
Clip:
Frame:
Error map:
Model 1:
Model 2:
Model 3:
Drag a red rectangle to the area that you want to see zoomed-in
Source
undef
undef
undef
2× Charts
Submitting
To add your upscaling method to MSU-VUB 2021, follow these steps:
1. Download input data |
Download the test video
Alternatively you can download it as png sequence |
2. Apply your upscaler |
Apply your 2× or 4× upscaling method to the downloaded video Configure it for lossless (png/tiff) output, if possible |
3. Send us the result |
Send video-upscalers-benchmark@videoprocessing.ai
|
See also
If you are interested in detail-preserving upscaling, check the MSU Video Super-Resolution Benchmark.
If you are interested in how compression with different video codec standards affects upscaling, check the MSU Super-Resolution for Video Compression Benchmark.
-
MSU Benchmark Collection
- MSU Super-Resolution for Video Compression Benchmark 2022
- MSU Video Quality Metrics Benchmark 2022
- MSU Video Upscalers Benchmark 2022
- MSU Video Alignment and Retrieval Benchmark
- MSU Mobile Video Codecs Benchmark 2021
- MSU Video Super-Resolution Benchmark
- MSU Shot Boundary Detection Benchmark 2020
- MSU Deinterlacer Benchmark
- The VideoMatting Project
- Video Completion
- Codecs Comparisons & Optimization
- VQMT
- Video Quality Measurement Tool 3D
-
MSU Datasets Collection
-
Metrics Research
- Video Filters
-
Other Projects