aiqbench analysis
VL-AcneSeg introduces a vision-language framework utilizing CLIP and region-level text prompts to improve acne lesion segmentation without requiring specific location data. The model demonstrates robust generalization on uncontrolled smartphone imagery, achieving performance comparable to expert annotations for severity estimation. This work highlights the potential of multimodal architectures to address niche medical imaging tasks through zero-shot adaptation.
VL-AcneSeg: A Vision-Language Framework for Region-Aware Acne Lesion Segmentation
arXiv:2609.34472v1 Announce Type: cross Abstract: Acne assessment is crucial for clinical decision-making, yet traditional grading and counting are subjective and fail to account for lesion size. While area-based assessment has emerged as a promising alternative, acne segmentation has continued to rely on general-purpose architectures. To address this gap, we propose VL-AcneSeg, 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. Because region-level prompts indicate which facial areas contain lesions, we report a single global prompt, which requires no such information, as our primary setting. On our internal clinical dataset, VL-AcneSeg achieves a Dice score of 0.5082 and an IoU of 0.3407 under this protocol, the highest among all compared methods, including recent vision-language segmentation methods that are themselves given region-level prompts; region-level prompting raises these to 0.5296 and 0.3602. Moreover, lesion area measurements derived from our segmentation correlate with IGA scores at a level comparable to expert annotations (Pearson r = 0.719 versus 0.658). Notably, our framework maintains consistent performance across external validation datasets, performing reliably even on uncontrolled smartphone images without requiring additional training or fine-tuning. By pairing a protocol that requires no lesion-location information with area-based severity estimation, this work provides a foundation for objective acne assessment outside the clinic. Our implementation is publicly available at: https://github.com/sukjuoh/VL-AcneSeg
Abstract:Acne assessment is crucial for clinical decision-making, yet traditional grading and counting are subjective and fail to account for lesion size. While area-based assessment has emerged as a promising alternative, acne segmentation has continued to rely on general-purpose architectures. To address this gap, we propose VL-AcneSeg, 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. Because region-level prompts indicate which facial areas contain lesions, we report a single global prompt, which requires no such information, as our primary setting. On our internal clinical dataset, VL-AcneSeg achieves a Dice score of 0.5082 and an IoU of 0.3407 under this protocol, the highest among all compared methods, including recent vision-language segmentation methods that are themselves given region-level prompts; region-level prompting raises these to 0.5296 and 0.3602. Moreover, lesion area measurements derived from our segmentation correlate with IGA scores at a level comparable to expert annotations (Pearson r = 0.719 versus 0.658). Notably, our framework maintains consistent performance across external validation datasets, performing reliably even on uncontrolled smartphone images without requiring additional training or fine-tuning. By pairing a protocol that requires no lesion-location information with area-based severity estimation, this work provides a foundation for objective acne assessment outside the clinic. Our implementation is publicly available at: this https URL
Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
arXiv:2609.34472 [cs.CV]
(or arXiv:2609.34472v1 [cs.CV] for this version)
https://doi.org/10.48550/arXiv.2609.34472 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 324 words
Related coverage
Med-OPD: Improving Medical Vision-Language Models via Evidence-Aware On-Policy Distillation
arXiv · Jul 21, 2026
StenoVLA-3D: 3D-Aware Reasoning VLA for Navigation Through Gastrointestinal Stenoses
arXiv · Sep 22, 2026
Semi-MedRef: Semi-Supervised Medical Referring Image Segmentation with Cross-Modal Alignment
arXiv · Aug 4, 2026
BUSTR: Descriptor-Aware Vision-Language Learning for Breast Ultrasound Report Generation
arXiv · Jul 21, 2026