LoViF 2026 Challenge on Human-oriented Semantic Image Quality Assessment: Methods and Results

2026-04-13

AI Summary

This work addresses the limitation of existing image quality assessment (IQA) methods, which predominantly focus on pixel-level distortions while neglecting human perception of semantic content. To bridge this gap, we introduce a novel IQA task explicitly oriented toward human semantic perception and present SeIQA, a large-scale, human-annotated dataset comprising training, validation, and test splits to enable end-to-end development and evaluation of semantic-aware IQA models. Leveraging this benchmark, we organized an international challenge that attracted 58 participating teams, six of which submitted valid solutions and established the current state-of-the-art performance on the test set. This effort not only sets a new standard for semantic IQA but also catalyzes emerging research directions such as semantic-aware image coding and optimization.

Technology Category

• Computer Vision: Segmentation
• Knowledge Representation and Reasoning: Qualitative Reasoning
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Semantics and Knowledge: Methods, algorithms and applications for the development of semantic models, knowledge graphs and other forms of structured data models with machine-interpretable semantics

Abstract

This paper reviews the LoViF 2026 Challenge on Human-oriented Semantic Image Quality Assessment. This challenge aims to raise a new direction, i.e., how to evaluate the loss of semantic information from the human perspective, intending to promote the development of some new directions, like semantic coding, processing, and semantic-oriented optimization, etc. Unlike existing datasets of quality assessment, we form a dataset of human-oriented semantic quality assessment, termed the SeIQA dataset. This dataset is divided into three parts for this competition: (i) training data: 510 pairs of degraded images and their corresponding ground truth references; (ii) validation data: 80 pairs of degraded images and their corresponding ground-truth references; (iii) testing data: 160 pairs of degraded images and their corresponding ground-truth references. The primary objective of this challenge is to establish a new and powerful benchmark for human-oriented semantic image quality assessment. There are a total of 58 teams registered in this competition, and 6 teams submitted valid solutions and fact sheets for the final testing phase. These submissions achieved state-of-the-art (SOTA) performance on the SeIQA dataset.

Problem

• semantic image quality assessment
• human-oriented evaluation
• semantic information loss
• image quality benchmark

Innovation

• semantic image quality assessment
• human-oriented evaluation
• SeIQA dataset
• semantic coding
• benchmark challenge