REBASE: Reference-Background Subspace Elimination for Training-Free In-Context Segmentation

• 类型：arxiv
• 标识：2607.09082
• 链接：https://arxiv.org/abs/2607.09082
• 主分类：engineering
• 形态：method
• 被引：0
• 被引来源：Semantic Scholar
• S2被引：0
• 影响力被引：0
• TLDR：This paper identifies the low-rank background feature subspace from the reference image and project the reference and query features onto its orthogonal complement in closed form, yielding cleaner semantic matching, and establishes a new state of the art among training-free methods on PACO-Part, FSS-1000, and cross-domain datasets such as ISIC2018, demonstrating that explicit background subspace removal is a highly effective principle for one-shot localization.
• 待LLM分类：否
• 标题中文：REBASE：参考-背景子空间消除的无训练上下文分割
• TLDR中文：本文从参考图像中识别低秩背景特征子空间，并以闭式方式将参考与查询特征投影到其正交补空间，从而获得更清晰的语义匹配，并在 PACO-Part、FSS-1000 以及 ISIC2018 等跨域数据集的无训练方法中达到新的 SOTA，表明显式去除背景子空间是一次性定位中极为有效的原则。
• 来源文件：
• /inbox/tom/_candidates/2026-07-21-rag-retrieval-reranking-candidates.json
• /inbox/tom/_candidates/2026-07-21-agent-rag-longcontext-candidates.json
• [S2 enrich]