Popular GitHub Repositories for Perceptually Degraded

RobustSAM: Segment Anything Robustly on Degraded Images (CVPR 2024 Highlight)

This project employs the GFPGAN algorithm to upscale and restore images. The tool leverages state-of-the-art deep learning models to enhance image quality and potentially restore degraded parts of the image.

RNN model to punctuate degraded text with no punctuation, and an application that combines it with Watson TTS for automated transcription.

Supplementary material for the SPICE-HL3 dataset

This project demonstrates a deep learning approach to enhance low-quality or degraded images using a neural network architecture.

This project focuses on developing perceptually-aware models for assessing the quality of filter-altered images.