MIEScore: Human-Aligned Evaluation for Multi-Source Image Editing
2026-08-03 • Computer Vision and Pattern Recognition
Computer Vision and Pattern RecognitionMultimedia
AI summaryⓘ
The authors created MIE-Bench, a new large-scale benchmark to better test how well AI models can edit images using multiple source images at once. They noticed that previous tests mostly focused on editing one image at a time and ignored the harder task of multi-source image editing. Their benchmark includes thousands of examples, human ratings, and uses 12 leading image editing models. They also developed MIEScore, an evaluation model that aligns closely with human opinions when judging edited images. This work helps improve and fairly evaluate tools that combine elements from several images based on text instructions.
multimodal modelsmulti-source image editingimage editing benchmarkhuman preference annotationmean opinion scoresinstruction followingattribute preservationmultimodal large language modelevaluation metricimage editing assessment
Authors
Zitong Xu, Huiyu Duan, Xinyun Zhang, Weifei Xiong, Tianyi Zheng, Xiongkuo Min, Qiang Hu, Zhengxue Cheng, Bo Li, Guangtao Zhai
Abstract
Recent advances in unified multimodal models have significantly improved text-guided image editing abilities. In particular, models such as Nano-Banana-Pro and GPT-Image-2 demonstrate emerging capabilities in multi-source image editing (MIE), including tasks such as object synthesis, person-background composition, and cross-image style fusion. However, existing benchmarks and image editing assessment (IEQA) methods remain primarily focused on single-image editing tasks and largely overlook the more challenging setting of MIE. This highlights the urgent need for a comprehensive and human-aligned benchmark for MIE. To this end, we introduce MIE-Bench, the first large-scale multiple image editing benchmark with fine-grained human preference annotations. Specifically, MIE-Bench includes 3,000 editing instances across 16 tasks, each involving more than two source images and an editing prompt, together with 36K edited images produced by 12 state-of-the-art editing models and over 108K mean opinion scores (MOSs) covering visual quality, instruction following, and attribute preservation. Based on MIE-Bench, we propose MIEScore, a multimodal large language model (MLLM)-based evaluation model enhanced with skill optimization and multi-dimensional supervised fine-tuning, to provide human-aligned feedback for MIE. Extensive experiments show that MIEScore achieves state-of-the-art performance in aligning with human preferences and generalizes well across other IEQA datasets. Both the dataset and the model are available at https://github.com/IntMeGroup/MIEScore.