MIRNet Low-Light Enhancement + ESRGAN Super-Resolution

This repository combines the power of MIRNet (Multi-Scale Residual Network) for low-light image enhancement and ESRGAN (Enhanced Super-Resolution GAN) for high-resolution image reconstruction. The pipeline first improves low-light images using MIRNet and then applies ESRGAN to upscale the resolution, making it ideal for enhancing and super-resolving dark and blurry images.

Features

• Low-Light Enhancement: Uses MIRNet to improve the visibility of images taken in poor lighting conditions.
• Super-Resolution: Utilizes ESRGAN to upscale the image, enhancing fine details and textures.
• End-to-End Pipeline: Input an image and get a fully enhanced high-resolution output in a few steps.

Key Highlights:

• Features section describes what the combined model does.
• Usage Just run the code and do one thing adjust the path of input output.
• Contributing invites users to participate in the project.
• License details the licensing information.

Let me know if you'd like to adjust any part!