D-Judge: How Far Are We? Evaluating the Discrepancies Between AI-synthesized Images and Natural Images through Multimodal Guidance
In Artificial Intelligence Generated Content (AIGC), distinguishing AI-synthesized images from natural ones remains a key challenge. Despite advancements in generative models, significant discrepancies persist. To systematically investigate and quantify these discrepancies, we introduce an AI-Natural Image Discrepancy accessing benchmark (\textit{D-Judge}) aimed at addressing the critical question: \textit{how far are AI-generated images (AIGIs) from truly realistic images?} We construct \textit{D-ANI}, a dataset with 5,000 natural images and over 440,000 AIGIs generated by nine models using Text-to-Image (T2I), Image-to-Image (I2I), and Text and Image-to-Image (TI2I) prompts. Our framework evaluates the discrepancy across five dimensions: naive image quality, semantic alignment, aesthetic appeal, downstream applicability, and human validation. Results reveal notable gaps, emphasizing the importance of aligning metrics with human judgment. Source code and datasets are available at https://shorturl.at/l83W2.
Code (1)
Tasks
multimodal generationSimilar Papers 제목 키워드 기반
PET Synthesis via Self-supervised Adaptive Residual Estimation Generative Adversarial Network
Positron emission tomography (PET) is a widely used, highly sensitive molecular imaging in clinical diagnosis. There is interest in reducing the radiation exposure from PET but also maintaining adequate image quality. Re…
Generative Adversarial NetworkPCSGAN: Perceptual Cyclic-Synthesized Generative Adversarial Networks for Thermal and NIR to Visible Image Transformation
In many real world scenarios, it is difficult to capture the images in the visible light spectrum (VIS) due to bad lighting conditions. However, the images can be captured in such scenarios using Near-Infrared (NIR) and …
Generative Adversarial NetworkSSIMCriterAlign: Criterion-Centric Rationale Alignment for Code Preference Judging
Pairwise human preference prediction is central to evaluating code-generation systems, where quality often depends on task-specific trade-offs beyond functional correctness. While rubric-based LLM judges improve interpre…
Controversial stimuli: pitting neural networks against each other as models of human recognition
Distinct scientific theories can make similar predictions. To adjudicate between theories, we must design experiments for which the theories make distinct predictions. Here we consider the problem of comparing deep neura…
Adversarial AttackBridging Human and LLM Judgments: Understanding and Narrowing the Gap
Large language models are increasingly used as judges (LLM-as-a-judge) to evaluate model outputs at scale, but their assessments often diverge systematically from human judgments. We present Bridge, a unified statistical…