sczhou/Upscale-A-Video

Health v2 · maintenance only

Flags: no_releases no_license

Adoption not part of the score

1470 stars · 82 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

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

library · maturity active

video-processing machine-learning deep-learning image-processing computer-vision deep-learning artificial-intelligence python video-super-resolution diffusion-model temporal-consistency deflicker cvpr-2024 aigc-enhancement research-code video linux gpu

Member repositories

For agents

markdown · JSON · MCP: product_card(name="sczhou/Upscale-A-Video")