VL-AcneSeg: A Vision-Language Framework for Region-Aware Acne Lesion Segmentation.
VLAcneSeg is proposed, a multimodal framework for acne lesion segmentation that leverages CLIP and region-level text prompts to incorporate spatial priors, enabling lesions to be localized across the whole face.
Sukju Oh, Soo Ick Cho, D. Suh et al.
· IEEE journal of biomedical a... · 0 citations