no-reference-image-quality-assessment

[CVPR2023] Blind Image Quality Assessment via Vision-Language Correspondence: A Multitask Learning Perspective

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

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

Official implementation for CVPR2023 Paper "Re-IQA : Unsupervised Learning for Image Quality Assessment in the Wild"

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

Non-local Modeling for Image Quality Assessment

Fast Adversarial CNN-based Perturbation Attack on no-reference image- and video-quality metrics

[TBC 24] Transformer-based Light Field Geometry Learning for No-reference Light Field Image Quality Assessment

Conditional Knowledge Distillation Network (CKDN) for blind image quality assessment of restored images using a degraded reference (PyTorch, ResNet-50 two-branch: QSE + DTE). Trained/evaluated on CSIQ and TID2013 with SRCC/PLCC metrics.

CEA Research Internship repository for feature-based No-Reference Underwater Image Quality Assessment (NR-UIQA), including MATLAB feature extraction, Python ensemble models, and experimental results.

A fast, no-reference video quality benchmarking tool using BRISQUE and other IQA metrics. Extracts sampled frames, computes perceptual quality scores, and compares encodes objectively.

No-reference image quality assessment extension for 3D slicer. Measure signal-to-noise ratio (SNR), contrast, contrast-to-noise ratio (CNR) and spatial resolution.