用目标检索提升SAM3在目标消失重现时的分割鲁棒性
Re-Prompting SAM 3 via Object Retrieval: 3rd of the 5th PVUW MOSE Track
- 基于DINOv3匹配候选目标，动态找回可靠锚点
- 多锚点传播使分割准确率达J&F 51.17%
- 适合处理目标遮挡、变形和同类干扰场景
本文探讨了PVUW 2026挑战赛MOSEv2赛道，聚焦复杂半监督视频对象分割。基于SAM~3，我们构建了一个自动重提示框架，以增强目标消失与重现、严重形变及同类别干扰下的鲁棒性。方法首先使用SAM~3检测后续帧中的同类别目标候选，再通过DINOv3实现基于对象级别的匹配，并结合变换感知的目标特征池进行可靠目标锚点检索。这些锚点与首帧掩码一同注入SAM~3追踪器，实现多锚点传播，而非仅依赖初始提示。该方法有效应对了MOSEv2的多个核心挑战。在测试集上，方案取得J&F 51.17%的性能，位列MOSEv2赛道第3名。
原文摘要 · Abstract (English)
This technical report explores the MOSEv2 track of the PVUW 2026 Challenge, which targets complex semi-supervised video object segmentation. Built on SAM~3, we develop an automatic re-prompting framework to improve robustness under target disappearance and reappearance, severe transformation, and strong same-category distractors. Our method first applies the SAM~3 detector to later frames to identify same-category object candidates, and then performs DINOv3-based object-level matching with a transformation-aware target feature pool to retrieve reliable target anchors. These anchors are injected back into the SAM~3 tracker together with the first-frame mask, enabling multi-anchor propagation rather than relying solely on the initial prompt. This simple directly benefits several core challenges of MOSEv2. Our solution achieves a J&F of 51.17% on the test set, ranking 3rd in the MOSEv2 track.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。