# NJU-PCALab/STAR

[ICCV 2025] STAR: Spatial-Temporal Augmentation with Text-to-Video Models for Real-World Video Super-Resolution

Repository: https://github.com/NJU-PCALab/STAR
Canonical: https://ross.abutalabs.com/products/nju-pcalab-star
Language: Python
License Family: other
Last push: 2025-07-02T08:32:10+00:00

## Health v2 (maintenance only)
Score: 34/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 29, release rhythm 35, longevity 46
- inputs: {"age_days": 650, "days_push": 427, "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 1495, forks 87 (observed 2026-08-28T04:04:53.059331+00:00)

## What it is
STAR is a research implementation of an ICCV 2025 paper performing real-world video super-resolution using spatial-temporal augmentation with text-to-video diffusion models. It provides inference and training code (I2VGen-XL version) plus online demos on HuggingFace and Colab.

## Use cases
- upscale low-quality real-world videos
- restore degraded video to high resolution
- enhance old compressed video footage
- run diffusion-based video super-resolution
- reproduce video restoration research results

## When to choose
- you need state-of-the-art real-world video super-resolution
- you want a diffusion-model-based video restoration pipeline with training code
- you want to build on published ICCV 2025 research

## When to avoid
- you need a lightweight real-time video upscaler
- you lack a GPU or cannot run large diffusion models
- you need a commercially licensed tool (no license is provided)

## Facets
- artifact type: library
- maturity: active
- function: video-processing, image-processing, deep-learning, stable-diffusion
- domain: computer-vision, machine-learning, image-processing
- platform: python
- tags: video-super-resolution, diffusion-models, text-to-video, iccv-2025, research-code, video, gpu, linux

## Member repositories
- NJU-PCALab/STAR (main) score 34

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:53.059331+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 2026-08-30T04:33:16.062085+00:00, confidence not recorded.
  - readme: https://github.com/NJU-PCALab/STAR (fetched 2026-08-28T04:04:53.059331+00:00, sha 99f7bb268d5a)
- Data as of 2026-08-30T08:39:29.467469+00:00.