image-degradation

Learning a Single Convolutional Super-Resolution Network for Multiple Degradations (CVPR, 2018) (Matlab)

[WACV 2024 Oral] - ARNIQA: Learning Distortion Manifold for Image Quality Assessment

Quality-Aware Image-Text Alignment for Opinion-Unaware Image Quality Assessment

(ECCV 2026) Official repository for MedQ-Deg, a benchmark evaluating medical VLM robustness to image degradation. Features 24,894 QA pairs across 7 imaging modalities, 18 degradation types, and 30 clinical capabilities, with multi-severity evaluation and confidence calibration analysis.

Python library for realistically degrading images.

[ICCV 2025] - Image Intrinsic Scale Assessment: Bridging the Gap Between Quality and Resolution

🔷 Effects of Degradations on Deep Neural Network Architectures (official code).

An AI restoration suite for cel animation, featuring specialized Real-ESRGAN models and the physics-based degradation simulator used to train them.

Can we perform face hallucination using limited set of unaligned pairs?

Using an already existing ESRGAN model, degrading our HR images to generate training data for finetuning and further checking the MSE and PSNR for the generated and the original HR images. Link to the original link of Real ESRGAN Repo - https://github.com/xinntao/Real-ESRGAN

A robust learning scheme by proposing a novel augmentation algorithm that scales to emulate a large set of image degradation and occlusion policies

Realistic synthetic degradation of document images — simulate scans, phone photos, photocopies, and aged paper for data augmentation in document AI / OCR / layout analysis pipelines. 23 physically-motivated transforms, 6 presets, CLI + Python API.

Digital Image Processing Course | Home Works Design| Fall 2021 | Dr. MohammadReza Mohammadi

Includes image smoothing, sharpening, and hough transfer.

🎨 Enhance cel animation with specialized AI models and a physics-based simulator for effective restoration of historical animation defects.

Evaluating face verification robustness under controlled image degradation using FaceNet embeddings and the LFW dataset.

Deep-learning project for robust AI-generated face detection, comparing degradation-aware training, expert routing and unified detection with PyTorch and ResNet-18.

Robust CNN & CapsuleNet benchmarks for noisy images – modern TensorFlow 2/Keras 3 support.

FaceTwin Project public dataset (CC BY 4.0) — three production prompts, six image-degradation profiles, sample-output metadata. Companion data for https://pleasejuststop.org and https://prankmyface.lol.

Reproducible research framework for evaluating vision-model reliability under controlled image degradation, with calibration, failure-detection and prediction-level analysis.