SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2203.15958 · CVPR · 2022

High-resolution Face Swapping via Latent Semantics Disentanglement

Jia Pan, Junle Wang, Shengfeng He, Yangyang Xu, Bailin Deng, Yanqing Jing

arXiv · PDF · Open in the Atlas

Code that ran

We have not lifted any functions out of this paper's repositories yet, so there is nothing we have run. The repositories linked to it are listed below.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Kong (a) Target (b) Source 1 (c) MegaFS [55] (d) Ours (e) Source 2 (f) MegaFS [55] (g) Ours Figure 1. We propose a high-resolution (1024×1024) face swapping method by disentangling the latent semantics of a pre-trained GAN model. Compared with the existing high-resolution method MegaFS [55], our model can preserve target attributes (e.g., skin color and illumination), meanwhile transferring local details and identity from source faces.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2203.15958")
get_code_for_paper("2203.15958")
have("2203.15958")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2203.15958.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2203.15958.svg)](https://syntology.ai/paper/2203.15958)

Connect an agent — have() is free.