CHIMERA: Controllable High-quality Image-Mask Extraction for Reliable Diffusion-based Anomaly Synthesis
JoungBin Lee, Hyunkoo Lee, Jini Yang, Chaehyun Kim, Jung Yi, Seok Hwangbo, Hyeoncheol Lee, Minho Chun
Abstract
We present CHIMERA, a novel framework for generating realistic, generalizable, and prompt-driven industrial anomalies from natural language instructions. Our method addresses two key challenges in text-guided anomaly synthesis: (1) the scarcity of scalable, high-quality paired anomaly data and (2) the difficulty of efficiently adapting large diffusion models to domain-specific tasks without overfitting. To tackle these challenges, we first introduce a Vision-Language Model (VLM)-guided data curation pipeline that automatically generates semantically rich and spatially grounded captions from normal images, enabling effective dataset augmentation without manual annotations. Building upon this, we propose a parameter-efficient fine-tuning strategy that adapts a pre-trained Diffusion Transformer (Stable Diffusion 3) using lightweight LoRA adapters. By aligning structured prompts with the model
BibTeX
@inproceedings{aaai2026_chimeracontrolla,
title = {CHIMERA: Controllable High-quality Image-Mask Extraction for Reliable Diffusion-based Anomaly Synthesis},
author = {JoungBin Lee and Hyunkoo Lee and Jini Yang and Chaehyun Kim and Jung Yi and Seok Hwangbo and Hyeoncheol Lee and Minho Chun and Eunjo Jeong and Seungryong Kim},
booktitle = {AAAI 2026},
year = {2026}
}