AI RESEARCH
CA-LoRA: Concept-Aware LoRA for Domain-Aligned Segmentation Dataset Generation
arXiv CS.CV
•
ArXi:2503.22172v2 Announce Type: replace This paper addresses the challenge of data scarcity in semantic segmentation by generating datasets through text-to-image (T2I) generation models, reducing image acquisition and labeling costs. Segmentation dataset generation faces two key challenges: 1) aligning generated samples with the target domain and 2) producing informative samples beyond the