# luanfujun/deep-painterly-harmonization

**Attribution required: if you use, quote, or summarise this content, you must credit and link back to [awesome-repositories.com](https://awesome-repositories.com/repository/luanfujun-deep-painterly-harmonization).**

6,043 stars · 610 forks · Cuda

## Links

- GitHub: https://github.com/luanfujun/deep-painterly-harmonization
- awesome-repositories: https://awesome-repositories.com/repository/luanfujun-deep-painterly-harmonization.md

## Description

Deep Painterly Harmonization is a deep learning image harmonization tool and convolutional neural network framework. It is designed to automate the process of blending the visual appearance of a foreground object into a background image to make composite images look natural.

The system functions as a computer vision blending tool that adjusts colors and lighting of inserted objects. It ensures that these elements match the global illumination and color palette of the target background environment.

The project covers digital image compositing and visual content integration by resolving color discrepancies between disparate image layers.

## Tags

### Graphics & Multimedia

- [Automated Blending](https://awesome-repositories.com/f/graphics-multimedia/media-processing-analysis/face-portrait-manipulation/image-blending/automated-blending.md) — Uses deep learning to automatically resolve color discrepancies when inserting new elements into photographs.
- [Image Color Harmonization](https://awesome-repositories.com/f/graphics-multimedia/image-color-harmonization.md) — Adjusts colors and lighting of inserted objects to blend naturally with the background environment. ([source](https://github.com/luanfujun/deep-painterly-harmonization#readme))
- [Image Compositing Replacements](https://awesome-repositories.com/f/graphics-multimedia/ml-background-replacements/image-compositing-replacements.md) — Places extracted foreground subjects onto different background images to produce a seamless composited output. ([source](https://github.com/luanfujun/deep-painterly-harmonization/blob/master/README2.md))
- [Image Composition](https://awesome-repositories.com/f/graphics-multimedia/image-editing-processing/image-processing/image-sequence-processors/animation-frame-sequencers/generative-animation-sequences/image-to-video-animators/image-composition.md) — Combines multiple image files into a single composite image by matching visual style and lighting.
- [Visual Content Integration](https://awesome-repositories.com/f/graphics-multimedia/visual-content-integration.md) — Ensures inserted objects match the global illumination and color palette of the target background image.

### Artificial Intelligence & ML

- [Color Matching Architectures](https://awesome-repositories.com/f/artificial-intelligence-ml/convolutional-neural-network-frameworks/color-matching-architectures.md) — Implements a convolutional neural network architecture designed to blend the visual appearance of foreground objects into backgrounds.
- [Deep Learning Image Processors](https://awesome-repositories.com/f/artificial-intelligence-ml/deep-learning-image-processors.md) — Utilizes deep neural networks to interpret image structures and automate shading and color harmonization tasks.
- [Computer Vision Research](https://awesome-repositories.com/f/artificial-intelligence-ml/computer-vision-research.md) — Provides a research implementation of painterly harmonization algorithms for visual analysis and image processing.

### User Interface & Experience

- [Image Color Extractors](https://awesome-repositories.com/f/user-interface-experience/color-palette-generators/image-color-extractors.md) — Adjusts the colors and lighting of foreground objects to harmonize with the background environment.