【图像增强】色彩和像素映射-High-Resolution Image Harmonization via Collaborative Dual Transformati-2023.04.06

目录

Target

Problem

This paper

Contribution

Method

Pixel to pixel

rgb to rgb

refined module

---

Resolution_Image_Harmonization_via_Collaborative_Dual_Transformations_CVPR_2022_paper

Target

image harmonization aims to adjust the foreground to make it compatible with the background.

Problem:

Conventional image harmonization methods learn global RGB-to-RGB transformation which could effort lessly scale to high resolution, but ignore diverse local context. Recent deep learning methods learn the dense pixel-to-pixel transformation which could generate harmonious outputs, but are highly constrained in low resolution.

This paper:

High-resolution image harmonization network with Collaborative Dual Transformation (CDTNet) to combine pixel-to-pixel transformation and RGB-to-RGB transformation coherently in an end-to-end network.

Contribution:

• We unify pixel-to-pixel transforma