Image Quality Assessment Comparison Hub
5 papers - avg viability 6.4
Top Papers
- EvoIQA - Explaining Image Distortions with Evolved White-Box Logic(7.0)
EvoIQA offers an explainable framework for image quality assessment using evolved mathematical formulas.
- Reference-Free Image Quality Assessment for Virtual Try-On via Human Feedback(7.0)
A reference-free framework for evaluating image quality in virtual try-on applications using human feedback.
- TIQA: Human-Aligned Text Quality Assessment in Generated Images(7.0)
TIQA is a new task and method for assessing the quality of text rendering in generated images, offering a practical solution for filtering and reranking in text-to-image pipelines.
- Self-Supervised Image Super-Resolution Quality Assessment based on Content-Free Multi-Model Oriented Representation Learning(6.0)
Develop a no-reference image quality assessment tool using self-supervised learning for super-resolution applications in real-world scenarios.
- From Global to Granular: Revealing IQA Model Performance via Correlation Surface(5.0)
A new method, GMC, provides fine-grained analysis of IQA model performance, surpassing traditional metrics with detailed 3D correlation mapping.