# sczhou/Upscale-A-Video

[CVPR 2024] Upscale-A-Video: Temporal-Consistent Diffusion Model for Real-World Video Super-Resolution

Repository: https://github.com/sczhou/Upscale-A-Video
Canonical: https://ross.abutalabs.com/products/upscale-a-video
Language: Python
License: NOASSERTION
License Family: other
Topics: deflicker, video-diffusion-model, video-super-resolution, aigc-enhancement
Last push: 2024-09-27T10:50:59+00:00

## Health v2 (maintenance only)
Score: 26/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 71
- inputs: {"age_days": 1007, "days_push": 705, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1470, forks 82 (observed 2026-08-28T04:04:49.146550+00:00)

## What it is
Upscale-A-Video is a diffusion-based model for real-world video super-resolution that takes low-resolution videos and text prompts as inputs. It is the official research code for a CVPR 2024 Highlight paper from S-Lab, Nanyang Technological University.

## Use cases
- upscale low-resolution videos to high resolution
- restore real-world degraded videos
- enhance video quality with text prompts
- reduce flicker in upscaled video frames
- run video super-resolution with a diffusion model
- research temporal-consistent video restoration

## When to choose
- you need diffusion-based video super-resolution with temporal consistency
- you want text-guided control over video restoration results
- you are reproducing CVPR 2024 research or building on its models
- you have a GPU and can run Python inference locally

## When to avoid
- you need fast real-time video upscaling on limited hardware
- you want a polished end-user application rather than research code
- you need a permissively licensed library for commercial products
- you only need simple image (not video) upscaling

## Facets
- artifact type: library
- maturity: active
- function: video-processing, machine-learning, deep-learning, image-processing
- domain: computer-vision, deep-learning, artificial-intelligence
- platform: python
- tags: video-super-resolution, diffusion-model, temporal-consistency, deflicker, cvpr-2024, aigc-enhancement, research-code, video, linux, gpu

## Member repositories
- sczhou/Upscale-A-Video (main) score 26

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:49.146550+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 2026-08-30T04:34:51.317576+00:00, confidence not recorded.
  - readme: https://github.com/sczhou/Upscale-A-Video (fetched 2026-08-28T04:04:49.146550+00:00, sha 4ab9945a3f74)
- Data as of 2026-08-30T08:39:29.467469+00:00.