# sczhou/ProPainter

[ICCV 2023] ProPainter: Improving Propagation and Transformer for Video Inpainting

Repository: https://github.com/sczhou/ProPainter
Canonical: https://ross.abutalabs.com/products/propainter
Homepage: https://shangchenzhou.com/projects/ProPainter/
Language: Python
License: NOASSERTION
License Family: other
Topics: object-removal, video-completion, video-inpainting, video-outpainting, watermark-removal
Last push: 2025-02-19T12:07:56+00:00

## Health v2 (maintenance only)
Score: 22/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 7, release rhythm 8, longevity 78
- inputs: {"age_days": 1097, "days_push": 560, "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 6916, forks 817 (observed 2026-08-28T04:09:51.369170+00:00)

## What it is
ProPainter is a PyTorch-based video inpainting model from ICCV 2023 that combines dual-domain propagation with a mask-guided sparse video Transformer. It removes objects, completes masked regions, and outpaints videos with state-of-the-art quality and memory-efficient inference.

## Use cases
- remove unwanted objects from a video
- fill in masked or missing regions in video frames
- expand the field of view of a video via outpainting
- erase watermarks from video footage
- repair corrupted or occluded areas in video sequences

## When to choose
- you need high-quality video inpainting with strong temporal consistency
- you want to remove objects or watermarks from clips programmatically
- you have a GPU and want a research-grade model with memory-efficient inference options

## When to avoid
- you need real-time inpainting on consumer hardware
- you only need simple image (not video) inpainting
- you require a permissive license for commercial use - the license is custom/non-standard

## Facets
- artifact type: library
- maturity: stable
- function: video-processing, image-processing, machine-learning, deep-learning
- domain: computer-vision, image-processing, artificial-intelligence
- platform: python, cross-platform
- tags: video-inpainting, object-removal, video-completion, video-outpainting, watermark-removal, iccv-2023, transformer, optical-flow, video, gpu, linux

## Member repositories
- sczhou/ProPainter (main) score 22

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:09:51.369170+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 2026-08-29T17:41:27.092210+00:00, confidence not recorded.
  - readme: https://github.com/sczhou/ProPainter (fetched 2026-08-28T04:09:51.369170+00:00, sha 33be9f9dbc89)
  - homepage: https://shangchenzhou.com/projects/ProPainter/ (fetched 2026-08-29T08:37:20.426924+00:00, sha 43eafe6443c6)
- Data as of 2026-08-30T08:39:29.467469+00:00.