Crucial-Diff: A Unified Diffusion Model for Crucial Image and Annotation Synthesis in Data-scarce Scenarios
It addresses data scarcity and imbalance for detection and segmentation tasks in specific domains, offering a domain-agnostic solution that is incremental over existing generative methods.
The paper tackles data scarcity in domains like medical imaging and autonomous driving by proposing Crucial-Diff, a unified diffusion model that synthesizes crucial training samples to target downstream model weaknesses, achieving results such as 83.63% pixel-level AP on MVTec and 81.64% mIoU on a polyp dataset.
The scarcity of data in various scenarios, such as medical, industry and autonomous driving, leads to model overfitting and dataset imbalance, thus hindering effective detection and segmentation performance. Existing studies employ the generative models to synthesize more training samples to mitigate data scarcity. However, these synthetic samples are repetitive or simplistic and fail to provide "crucial information" that targets the downstream model's weaknesses. Additionally, these methods typically require separate training for different objects, leading to computational inefficiencies. To address these issues, we propose Crucial-Diff, a domain-agnostic framework designed to synthesize crucial samples. Our method integrates two key modules. The Scene Agnostic Feature Extractor (SAFE) utilizes a unified feature extractor to capture target information. The Weakness Aware Sample Miner (WASM) generates hard-to-detect samples using feedback from the detection results of downstream model, which is then fused with the output of SAFE module. Together, our Crucial-Diff framework generates diverse, high-quality training data, achieving a pixel-level AP of 83.63% and an F1-MAX of 78.12% on MVTec. On polyp dataset, Crucial-Diff reaches an mIoU of 81.64% and an mDice of 87.69%. Code is publicly available at https://github.com/JJessicaYao/Crucial-diff.