CLASP: Adaptive Spectral Clustering for Unsupervised Per-Image Segmentation

Max Curie, Paulo da Costa; Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV) Workshops, 2025, pp. 6041-6050

We introduce CLASP (Clustering via Adaptive Spectral Processing), a lightweight framework for unsupervised image segmentation that operates without any labeled data or fine-tuning. CLASP first extracts per-patch features using a self-supervised ViT encoder (DINO); then, it builds an affinity matrix and applies spectral clustering. To avoid manual tuning, we select the segment count automatically with a eigengap-silhouette search, and we sharpen the boundaries with a fully connected DenseCRF. Despite its simplicity and training-free nature, CLASP attains competitive mIoU and pixel-accuracy on COCO-Stuff and ADE20K, matching recent unsupervised baselines. The zero-training design makes CLASP a strong, easily reproducible baseline for large unannotated corpora--especially common in digital advertising and marketing workflows such as brand-safety screening, creative asset curation, and social-media content moderation.