# JingyunLiang/VRT

VRT: A Video Restoration Transformer (official repository)

Repository: https://github.com/JingyunLiang/VRT
Canonical: https://ross.abutalabs.com/products/vrt
Homepage: https://arxiv.org/abs/2201.12288
Language: Python
License: NOASSERTION
License Family: other
Topics: transformer, video-restoration, low-level-vision, vision-transformer, video-super-resolution, video-deblurring, video-denoising, video-sr, super-resolution, sr, deblurring, denoising, video, restoration
Last push: 2023-06-18T18:07:39+00:00

## Health v2 (maintenance only)
Score: 23/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 8, longevity 100
- inputs: {"age_days": 1688, "days_push": 1172, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1546, forks 141 (observed 2026-08-28T04:05:01.636838+00:00)

## What it is
VRT is the official PyTorch implementation of the paper 'VRT: A Video Restoration Transformer', a transformer-based model for video restoration tasks. It provides pretrained models and training/evaluation code for video super-resolution, deblurring, denoising, and frame interpolation.

## Use cases
- restore low-quality video frames to high quality
- upscale low-resolution videos with super-resolution
- remove blur from shaky or motion-blurred videos
- denoise noisy video footage
- interpolate missing frames in a video sequence
- reproduce state-of-the-art video restoration benchmarks like REDS and Vimeo90K

## When to choose
- you need a transformer-based model for video super-resolution, deblurring, or denoising
- you want pretrained checkpoints to run video restoration experiments quickly
- you are doing research on low-level vision and need a strong baseline

## When to avoid
- you need real-time video processing on consumer hardware
- you want a production-ready application with a GUI rather than research code
- you need a permissively licensed library since the license is non-standard

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, deep-learning, image-processing, video-processing
- domain: computer-vision, deep-learning, machine-learning
- platform: python
- tags: vision-transformer, video-super-resolution, video-deblurring, video-denoising, frame-interpolation, pytorch, research-code, pretrained-models, video, gpu

## Member repositories
- JingyunLiang/VRT (main) score 23

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:01.636838+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T04:30:32.652848+00:00, confidence not recorded.
  - readme: https://github.com/JingyunLiang/VRT (fetched 2026-08-28T04:05:01.636838+00:00, sha a798db323f7a)
  - homepage: https://arxiv.org/abs/2201.12288 (fetched 2026-08-29T11:31:49.743204+00:00, sha 372b663884fc)
  - site_page: https://info.arxiv.org/about/donate.html (fetched 2026-08-29T11:31:49.752479+00:00, sha cca9c3a11c56)
  - site_page: https://info.arxiv.org/about/ourmembers.html (fetched 2026-08-29T11:31:49.756580+00:00, sha 47cbc55ff1de)
  - site_page: https://info.arxiv.org/about (fetched 2026-08-29T11:31:49.759733+00:00, sha a1f16f915a9a)
  - site_page: https://info.arxiv.org/labs/index.html (fetched 2026-08-29T11:31:49.754691+00:00, sha b14a8d05a0ec)
- Data as of 2026-08-30T08:39:29.467469+00:00.