# ai-forever/ghost

A new one shot face swap approach for image and video domains

Repository: https://github.com/ai-forever/ghost
Canonical: https://ross.abutalabs.com/products/ai-forever-ghost
Language: Python
License: Apache-2.0
License Family: permissive
Topics: computer-vision, deep-learning, deepfake, faceswap, pytorch, deep-face-swap, face-swap, ghost, ghost-faceswap, ghost-swap
Last push: 2025-02-25T10:15:05+00:00

## Health v2 (maintenance only)
Score: 26/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 8, release rhythm 8, longevity 100
- inputs: {"age_days": 1724, "days_push": 554, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1582, forks 308 (observed 2026-08-28T04:05:07.163258+00:00)

## What it is
GHOST (Generative High-fidelity One Shot Transfer) is a one-shot face swap pipeline for images and videos, published as an IEEE paper and implemented in PyTorch. It swaps a face from a single source image onto target images or videos using generative techniques.

## Use cases
- swap a face from one photo onto another image
- put a face onto a video with a single reference photo
- experiment with one-shot deepfake generation models
- reproduce GAN-based face swap research from the GHOST paper
- try face swapping in Google Colab without local GPU

## When to choose
- you need one-shot face swapping without training a model per identity
- you want a research-grade face swap implementation with a published paper
- you want to run face swap on both images and videos
- you want a Colab demo to test quickly

## When to avoid
- you need production-grade, polished end-user face swap software
- you require real-time face swapping performance
- your use case involves swapping faces without consent or other unethical purposes, which the project explicitly forbids
- you need a maintained GUI application rather than a research codebase

## Facets
- artifact type: library
- maturity: active
- function: computer-vision, image-processing, video-processing, deep-learning, machine-learning
- domain: computer-vision, deep-learning, image-processing
- platform: python, cross-platform
- tags: face-swap, deepfake, one-shot, gan, pytorch, research, video, gpu

## Member repositories
- ai-forever/ghost (main) score 26

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:07.163258+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 2026-08-30T03:56:08.268187+00:00, confidence not recorded.
  - readme: https://github.com/ai-forever/ghost (fetched 2026-08-28T04:05:07.163258+00:00, sha 0a86fab58302)
- Data as of 2026-08-30T08:39:29.467469+00:00.