D3S2: Diffusion-Guided Dataset Distillation for Semantic Segmentation
Researchers have developed D3S2, a novel framework for dataset distillation specifically designed for semantic segmentation tasks. This method addresses challenges like class imbalance and the need for precise pixel alignment by using a two-stage approach involving balanced mask selection and diffusion-guided image synthesis. D3S2 ensures spatial alignment and enhances training utility through guided diffusion sampling, achieving significant improvements in mean Intersection over Union (mIoU) scores on benchmark datasets even at a 1% compression rate. AI
IMPACT Introduces a new technique for creating smaller, effective datasets for complex image analysis tasks.