SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2204.13962 · CVPR · 2022

SCS-Co: Self-Consistent Style Contrastive Learning for Image Harmonization

Bin Xia, Wenming Yang, Qingmin Liao, Yucheng Hang

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

edu.cn (a) Composite (b) DoveNet [5] (c) iS 2 AM [35] (d) RainNet [24] (e) Ours (f) Real (Ground Truth) Figure 1. Comparison with other methods. Thanks to the proposed SCS-Co, our method can obtain more explicit distortion knowledge from dynamically generated negative samples, and further jointly constrain the solution space from two aspects of the foreground self-style and foreground-background style consistency. Coupled with BAIN, our method produces a more photorealistic harmonized result.

For agents

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

get_harvested_code_for_paper("2204.13962")
get_code_for_paper("2204.13962")
have("2204.13962")

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

curl https://syntology.ai/api/ran/2204.13962.json

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

[![Syntology run record](https://syntology.ai/api/ran/2204.13962.svg)](https://syntology.ai/paper/2204.13962)

Connect an agent — have() is free.