# svip-lab/impersonator

PyTorch implementation of our ICCV 2019 paper:  Liquid Warping GAN: A Unified Framework for Human Motion Imitation, Appearance Transfer and Novel View Synthesis

Repository: https://github.com/svip-lab/impersonator
Canonical: https://ross.abutalabs.com/products/impersonator
Homepage: https://svip-lab.github.io/project/impersonator
Language: Python
License: NOASSERTION
License Family: other
Topics: pytorch, pose, gan
Last push: 2022-06-21T22:56:20+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 2621, "days_push": 1534, "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 1717, forks 308 (observed 2026-08-28T04:05:26.399245+00:00)

## What it is
A PyTorch implementation of Liquid Gating GAN (ICCV 2019) that performs human motion imitation, appearance transfer, and novel view synthesis in a unified framework. It uses 3D body mesh recovery to disentangle pose and shape, plus a Liquid Warping Block to preserve source identity, texture, and style in generated images.

## Use cases
- make a person in an image follow a motion from a video
- transfer clothing appearance between people in photos
- generate novel views of a person from a single image
- animate a still photo with a driving pose sequence
- research human motion transfer with GANs
- evaluate motion imitation on the iPER dataset

## When to choose
- you need research-grade human motion imitation or appearance transfer in PyTorch
- you want a unified model handling motion, appearance, and view synthesis
- you want to reproduce or build on the ICCV 2019 Liquid Warping GAN paper
- you have a CUDA GPU with at least ~4GB memory

## When to avoid
- you need a production-ready, actively maintained product
- you work on Windows or macOS without testing effort
- you need real-time or mobile inference
- you want a simple API without downloading checkpoints and building thirdparty renderers

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, deep-learning, image-processing, computer-vision
- domain: deep-learning, computer-vision, image-processing
- platform: python
- tags: gan, pytorch, motion-imitation, appearance-transfer, novel-view-synthesis, human-pose, research-code, iccv-2019, linux, gpu

## Member repositories
- svip-lab/impersonator (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:26.399245+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 2026-08-30T03:34:14.123779+00:00, confidence not recorded.
  - readme: https://github.com/svip-lab/impersonator (fetched 2026-08-28T04:05:26.399245+00:00, sha 9e24910367f2)
  - homepage: https://svip-lab.github.io/project/impersonator (fetched 2026-08-29T11:10:13.745770+00:00, sha be4883bc7dbe)
- Data as of 2026-08-30T08:39:29.467469+00:00.