Metraipolis › archive › dc_c29a42d53d8a
SAFe: Segment-guided Aggregation of Feature Densities for Anomaly-aware Segmentation — arXiv:2609.24204
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.24204
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-09T22:05:22.469Z
- SHA-256 of the archived text
410c97dd44f13a7d2e9d0461e79707579939daa853698d8e8a8269bfcab6d5a5- Stable URL of this page
https://metraipolis.com/documento/dc_c29a42d53d8a
How to cite
SAFe: Segment-guided Aggregation of Feature Densities for Anomaly-aware Segmentation — arXiv:2609.24204 — arXiv (en), https://arxiv.org/abs/2609.24204 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_c29a42d53d8a (Metraipolis archive dc_c29a42d53d8a, sha256 410c97dd44f13a7d2e9d0461e79707579939daa853698d8e8a8269bfcab6d5a5)
BibTeX
@misc{dcc29a42d53d8a,
title = {SAFe: Segment-guided Aggregation of Feature Densities for Anomaly-aware Segmentation — arXiv:2609.24204},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.24204},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_c29a42d53d8a; sha256 410c97dd44f13a7d2e9d0461e79707579939daa853698d8e8a8269bfcab6d5a5},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.24204. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Visual segmentation systems encounter objects outside their training distribution during real-world deployment, hindering reliable autonomous systems that depend on scene parsing in the perception stage. Many recent methods address this by using self-supervised foundation models to train density estimators that yield low likelihood in anomalous image regions. Although promising, these methods suffer from poor feature semantics or they lack spatial consistency, both of which undermine critical downstream decisions. We address this problem with~\method, a generative method based on class-conditional density estimation over self-supervised representations. SAFe trains lightweight normalizing flows that produce class-conditional normalized likelihood estimates over frozen DINOv3 features. We combine density estimates from transformer features with density scores over multi-scale convolutional features to capture both global semantics and local detail. We introduce a method-agnostic post-processing step based on SAM3 that connects per-location likelihoods into spatially coherent segments while suppressing false positives, and enables instance-level anomaly detection without retraining. The post processing further distinguishes novel categories among anomalous objects by a similarity-based agglomerative clustering scheme. SAFe sets a new state of the art on the PANIC, OoDIS, SMIYC ObstacleTrack with strong performance on the ISSU benchmark. Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-22 in cs.CV). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV). Pagina del lavoro: https://arxiv.org/abs/2609.24204 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-09T21:45:21.006Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-09-22?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).