paper-with-me

홈 › Papers

Exposing Text-Image Inconsistency Using Diffusion Models

2024-04-28 · Mingzhen Huang, Shan Jia, Zhou Zhou, Yan Ju, Jialing Cai, Siwei Lyu

In the battle against widespread online misinformation, a growing problem is text-image inconsistency, where images are misleadingly paired with texts with different intent or meaning. Existing classification-based methods for text-image inconsistency can identify contextual inconsistencies but fail to provide explainable justifications for their decisions that humans can understand. Although more nuanced, human evaluation is impractical at scale and susceptible to errors. To address these limitations, this study introduces D-TIIL (Diffusion-based Text-Image Inconsistency Localization), which employs text-to-image diffusion models to localize semantic inconsistencies in text and image pairs. These models, trained on large-scale datasets act as ``omniscient" agents that filter out irrelevant information and incorporate background knowledge to identify inconsistencies. In addition, D-TIIL uses text embeddings and modified image regions to visualize these inconsistencies. To evaluate D-TIIL's efficacy, we introduce a new TIIL dataset containing 14K consistent and inconsistent text-image pairs. Unlike existing datasets, TIIL enables assessment at the level of individual words and image regions and is carefully designed to represent various inconsistencies. D-TIIL offers a scalable and evidence-based approach to identifying and localizing text-image inconsistency, providing a robust framework for future research combating misinformation.

📄 PDF Abstract BibTeX arXiv:2404.18033

Code (1)

Mingzhen-Huang/D-TIIL 공식 구현 pytorch

Tasks

Misinformation

Methods 이 논문이 사용한 방법론

Diffusion Diffusion models generate samples by gradually removing noise from a signal, and their training objective can be expressed as a reweighted variational lower-bound…

Similar Papers 제목 키워드 기반

Rethinking the Spatial Inconsistency in Classifier-Free Diffusion Guidance

2024-04-08 · CVPR 2024 1 · Dazhong Shen, Guanglu Song, Zeyue Xue, Fu-Yun Wang 외

Classifier-Free Guidance (CFG) has been widely used in text-to-image diffusion models, where the CFG scale is introduced to control the strength of text guidance on the whole image space. However, we argue that a global …

DenoisingSemantic Segmentation

Exposing the Fake: Effective Diffusion-Generated Images Detection

2023-07-12 · RuiPeng Ma, Jinhao Duan, Fei Kong, Xiaoshuang Shi 외

Image synthesis has seen significant advancements with the advent of diffusion-based generative models like Denoising Diffusion Probabilistic Models (DDPM) and text-to-image diffusion models. Despite their efficacy, ther…

DenoisingImage Generation

Prompt-Consistency Image Generation (PCIG): A Unified Framework Integrating LLMs, Knowledge Graphs, and Controllable Diffusion Models

2024-06-24 · Yichen Sun, Zhixuan Chu, Zhan Qin, Kui Ren

The rapid advancement of Text-to-Image(T2I) generative models has enabled the synthesis of high-quality images guided by textual descriptions. Despite this significant progress, these models are often susceptible in gene…

HallucinationImage GenerationKnowledge GraphsText Generation

MV2UV: Generating High-quality UV Texture Maps with Multiview Prompts

2026-03-16 · Zheng Zhang, Qinchuan Zhang, Yuteng Ye, Zhi Chen 외 arxiv

Generating high-quality textures for 3D assets is a challenging task. Existing multiview texture generation methods suffer from the multiview inconsistency and missing textures on unseen parts, while UV inpainting textur…

UPGPT: Universal Diffusion Model for Person Image Generation, Editing and Pose Transfer

2023-04-18 · Soon Yau Cheong, Armin Mustafa, Andrew Gilbert

Text-to-image models (T2I) such as StableDiffusion have been used to generate high quality images of people. However, due to the random nature of the generation process, the person has a different appearance e.g. pose, f…

DisentanglementImage GenerationPose TransferText-to-Image Generation+1