Datasets › MSU Video Upscalers: Quality Enhancement

MSU Video Upscalers: Quality Enhancement

archive 2025-07-28

The dataset aims to find the algorithms that produce the most visually pleasant image possible and generalize well to a broad range of content. It consists of 30 clips and contains 15 2D-animated segments losslessly recorded from various video games and 15 camera-shot segments from high-bitrate YUV444 sources. The complexity of clips varies significantly in terms of spatial and temporal indexes. Multiple bicubic downscaling mixed with sharpening is used to simulate complex real-world camera degradation. The authors used slight compression and YUV420 conversion to simulate a practical use case. 1920×1080 sources were downscaled to 480×270 input.

Benchmarks archive 2025-07-28

All 1 leaderboard whose dataset resolves to this page shown (sort by any header). "First row" is the archive's own first row at snapshot, in the archive's row order; nothing here re-ranks and metric direction is not asserted.

First row (archive order)PaperCode
Video Super-Resolution MSU Video Upscalers: Quality Enhancement BSRGAN LPIPS 0.177 Designing a Practical Degradation Model for Deep Blind... cszn/BSRGAN +2 48 Compare

Papers archive 2025-07-28

24 shown of 24 papers with a leaderboard row on this dataset's benchmarks, newest first. The archive's own "papers using this dataset" list was never published, so this is the benchmark-backed subset; the archive's count for this dataset is 24. The Syntology column is from Syntology's graph (read 2026-09-24), stated per sample; it is not part of any archive number.

DateSamples run Syntology
VRT: A Video Restoration Transformer 1 1 28 Jan 2022 ran 4 of 5 samples (1 unverified; 5 pointer-only for licence)
Investigating Tradeoffs in Real-World Video Super-Resolution 1 1 24 Nov 2021 not harvested
SwinIR: Image Restoration Using Swin Transformer 9 2 23 Aug 2021 ran 30 of 45 samples (15 unverified; 5 pointer-only for licence)
Real-ESRGAN: Training Real-World Blind Super-Resolution with Pure Synthetic Data 8 6 22 Jul 2021 ran 3 of 9 samples (6 unverified)
Real-Time Super-Resolution System of 4K-Video Based on Deep Learning 1 1 12 Jul 2021 not harvested
COMISR: Compression-Informed Video Super-Resolution 2 1 4 May 2021 ran 0 of 3 samples (3 unverified)
BasicVSR++: Improving Video Super-Resolution with Enhanced Propagation and Alignment 3 1 27 Apr 2021 not harvested
Designing a Practical Degradation Model for Deep Blind Image Super-Resolution 3 2 25 Mar 2021 ran 7 of 12 samples (5 unverified)
DynaVSR: Dynamic Adaptive Blind Video Super-Resolution 1 1 9 Nov 2020 not harvested
Local-Global Fusion Network for Video Super-Resolution 1 1 22 Sep 2020 not harvested
iSeeBetter: Spatio-Temporal Video Super Resolution using Recurrent-Generative Back-Projection Networks 1 1 21 Jun 2020 not harvested
Real-World Super-Resolution via Kernel Estimation and Noise Injection 2 1 19 Jun 2020 not harvested
Deep Blind Video Super-resolution 2 1 10 Mar 2020 ran 9 of 15 samples (6 unverified)
Deep Video Super-Resolution using HR Optical Flow Estimation 2 1 6 Jan 2020 not harvested
Learning Temporal Coherence via Self-Supervision for GAN-based Video Generation 13 1 23 Nov 2018 ran 2 of 25 samples (23 unverified)
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks 46 1 1 Sep 2018 ran 8 of 44 samples (36 unverified)
Frame-Recurrent Video Super-Resolution 0 1 14 Jan 2018 not harvested
Learning a Single Convolutional Super-Resolution Network for Multiple Degradations 1 1 17 Dec 2017 not harvested
Image Super-Resolution via Deep Recursive Residual Network 1 1 1 Jul 2017 not harvested
Detail-revealing Deep Video Super-resolution 1 1 10 Apr 2017 not harvested
Real-Time Video Super-Resolution with Spatio-Temporal Networks and Motion Compensation 0 1 16 Nov 2016 not harvested
Real-Time Single Image and Video Super-Resolution Using an Efficient Sub-Pixel Convolutional Neural Network 47 1 16 Sep 2016 ran 4 of 20 samples (16 unverified; 1 pointer-only for licence)
Accurate Image Super-Resolution Using Very Deep Convolutional Networks 8 1 14 Nov 2015 ran 0 of 6 samples (6 unverified)
Image Super-Resolution Using Deep Convolutional Networks 60 1 31 Dec 2014 ran 7 of 27 samples (20 unverified; 7 pointer-only for licence)

Dataset loaders archive 2025-07-28

No loader listed in the archive.

Tasks archive 2025-07-28

License archive 2025-07-28

No licence recorded in the archive. Absence here is not a statement about the dataset's terms.

Modalities archive 2025-07-28

No modality tagged.

Languages archive 2025-07-28

No language tagged.

Variants archive 2025-07-28

  • MSU Video Upscalers: Quality Enhancement

1 variant name, as the archive lists them.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections