← 研踪 PaperTrailarXiv:2406.10891 · v4
Sfumato Supervision: Learning Instance Segmentation with Noisy Labels
首次提交 2024-06-16 18:49
版本更新 2026-11-16 17:44
雷达观测 2026-11-17 17:11
摘要
Instance segmentation requires precise object masks, yet both human annotators and promptable models can produce unreliable contours. A high prediction loss near a contour does not necessarily indicate an annotation error: valid edges can also be uncertain. Inspired by Leonardo da Vinci's \emph{sfumato}, which softens transitions between a subject and its surroundings, we introduce \textbf{\texttt{CB-LNL} (Copula-Boundary Learning with Noisy Labels)}. Rather than judging each contour pixel alone, CB-LNL combines foreground evidence inside the mask with background evidence outside it to guide local supervision and annotation repair. We ground this approach in an audit of paired COCO and LVIS annotations, and introduce benchmarks covering human annotations, weak-prompt SAM2 masks, and controlled corruptions in natural and driving scenes. Controlled comparisons show improved training on procedural contour noise and fewer harmful edits than direct teacher-mask replacement. Paired-seed and frozen-prediction evaluations reveal how these benefits depend on the annotation source, supporting separate assessment of category and contour reliability.
观测到的版本
相邻记录可能跳过中间版本;首次发现 v2 及以上也会标为版本更新。
v4 · 2026-11-16 17:44 · 最新观测版本
Sfumato Supervision: Learning Instance Segmentation with Noisy Labels
Instance segmentation requires precise object masks, yet both human annotators and promptable models can produce unreliable contours. A high prediction loss near a contour does not necessarily indicate an annotation error: valid edges can also be uncertain. Inspired by Leonardo da Vinci's \emph{sfumato}, which softens transitions between a subject and its surroundings, we introduce \textbf{\texttt{CB-LNL} (Copula-Boundary Learning with Noisy Labels)}. Rather than judging each contour pixel alone, CB-LNL combines foreground evidence inside the mask with background evidence outside it to guide local supervision and annotation repair. We ground this approach in an audit of paired COCO and LVIS annotations, and introduce benchmarks covering human annotations, weak-prompt SAM2 masks, and controlled corruptions in natural and driving scenes. Controlled comparisons show improved training on procedural contour noise and fewer harmful edits than direct teacher-mask replacement. Paired-seed and frozen-prediction evaluations reveal how these benefits depend on the annotation source, supporting separate assessment of category and contour reliability.
查看此版本解读 →查看此版本原文 ↗