SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper ยท 2303.11545 ยท CVPR ยท 2023

Fix the Noise: Disentangling Source Feature for Controllable Domain Translation

Jaejun Yoo, Kaist, Jae Lee, Junmo Kim, Jaehyun Choi, Doyeon Kim, Dongyeun Lee, ๐ง๐ง ๐‘Ž๐‘Ž๐‘Ž๐‘Ž๐‘Ž๐‘Ž๐‘Ž, ๐ง๐ง ๐‘Ÿ๐‘Ÿ๐‘Ž๐‘Ž๐‘Ž๐‘Ž๐‘Ÿ๐‘Ÿ, โ„‹ ๐‘ก๐‘ก

arXiv ยท PDF ยท 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

Figure 1. Given a source model Gs, our method can smoothly control the degree of source domain features in a fine-tuned model Gsโ†’t. Samples in each row are generated from the same latent code z โˆˆ Z by Gs and Gsโ†’t. Here, Hs, Hsโ†’t, and Ht denote feature spaces of the source, our model, and simply fine-tuned target model, respectively. Our approach explicitly guides the model to preserve the source features by using the anchor point n anch , which allows a flexible and smooth cross-domain control via Gsโ†’t.

For agents

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

get_harvested_code_for_paper("2303.11545")
get_code_for_paper("2303.11545")
have("2303.11545")

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

curl https://syntology.ai/api/ran/2303.11545.json

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

[![Syntology run record](https://syntology.ai/api/ran/2303.11545.svg)](https://syntology.ai/paper/2303.11545)

Connect an agent โ€” have() is free.