Acknowledgments

Our AI Image Enhancer is powered by cutting-edge open-source AI models developed by leading researchers and organizations. We are deeply grateful to these contributors for their groundbreaking work and commitment to open science.

Built with 6 Open-Source AI Models

All models are used in compliance with their respective open-source licenses. Full attribution and license texts are provided below.

🔍

Real-ESRGAN

Version: x2plus, x4plus

BSD 3-Clause License

Purpose

AI-powered image super-resolution for 2x, 4x, and 8x upscaling

Authors

Xintao Wang, Liangbin Xie, Chao Dong, Ying Shan

Organization

Tencent ARC Lab

👤

GFPGAN

Version: v1.3, v1.4

Apache License 2.0

Purpose

Generative Facial Prior GAN for face restoration and enhancement

Authors

Xintao Wang, Yu Li, Honglun Zhang, Ying Shan

Organization

Tencent ARC Lab

✂️

BiRefNet

Version: Latest

MIT License

Purpose

Bilateral reference network for high-quality background removal

Authors

Zheng Peng

Organization

Independent Research

🎨

LaMa

Version: Resolution-robust Large Mask Inpainting

Apache License 2.0

Purpose

Large mask inpainting for object removal and image restoration

Authors

Roman Suvorov, Elizaveta Logacheva, Anton Mashikhin, et al.

Organization

Samsung AI Center

🔍

NAFNet

Version: REDS deblurring model

MIT License

Purpose

Nonlinear Activation Free Network for high-quality image deblurring

Authors

Liangyu Chen, Xiaojie Chu, Xiangyu Zhang, Jian Sun

Organization

Megvii Research

📝

EasyOCR

Version: Latest

Apache License 2.0

Purpose

Optical character recognition for text detection and extraction

Authors

Jaided AI Team

Organization

Jaided AI

If you use our service in your research or project, please consider citing the original papers of the AI models we use. This helps support the researchers and encourages continued innovation in open-source AI.