InterRVOS：交互感知的指代视频目标分割
InterRVOS: Interaction-Aware Referring Video Object Segmentation
将交互关系显式建模引入指代视频目标分割，要求分别分割动作执行者与目标对象，并设计了交互感知特殊 token 与注意力掩码损失；对做视频理解、指代分割和多模态推理的研究者，提供了可复用的角色区分分割范式与配套数据集。
原文依据 requiring separate segmentation of actor and target objects
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论指代视频目标分割（RVOS）旨在分割视频中由自然语言表达式描述的对象。然而，现有方法大多只关注被指代的对象（通常是动作执行者），即使表达式明确描述了涉及多个具有不同角色的对象的交互。本文提出交互感知的指代视频目标分割（InterRVOS），这是一个新任务，聚焦于显式交互建模，要求分别分割动作执行者与目标对象。这种形式化使得对对象关系的细粒度理解成为可能，因为许多视频事件是由此类交互而非单个对象定义的。我们提出了 InterRVOS-127K，一个包含超过 127K 条自动标注表达式的大规模数据集，每条表达式具有不同的动作执行者-目标掩码对，并提出 ReVIOSa，一种基于多模态大语言模型的架构，引入交互感知特殊 token 和注意力掩码损失（AML）以增强交互感知分割。我们还提出了一种新的评测协议，分别评估动作执行者和目标的分割，以实现更准确的角色区分。综合实验表明，ReVIOSa 在所提出的 InterRVOS-127K 基准上优于现有基线，进一步的分析验证了 ReVIOSa 和 InterRVOS-127K 的必要性与有效性。
英文原文摘要 1363 字符
Referring video object segmentation (RVOS) aims to segment objects in a video described by a natural language expression. However, most existing approaches focus only on the referred object (typically the actor), even when the expression clearly describes an interaction involving multiple objects with distinct roles. In this paper, we introduce Interaction-Aware Referring Video Object Segmentation (InterRVOS), a novel task that focuses on explicit interaction modeling by requiring separate segmentation of actor and target objects.This formulation enables fine-grained understanding of object relationships, as many video events are defined by such interactions rather than individual objects. We present InterRVOS-127K, a large-scale dataset of over 127K automatically annotated expressions with distinct actor-target mask pairs, and propose ReVIOSa, a MLLM-based architecture that introduces interaction-aware special tokens and attention mask loss (AML) to enhance interaction-aware segmentation. We also propose a new evaluation protocol that separately evaluates actor and target segmentation for more accurate role distinction. Comprehensive experiments demonstrate that ReVIOSa outperforms existing baselines on the proposed InterRVOS-127K benchmark, with further analyses validating the necessity and effectiveness of both ReVIOSa and InterRVOS-127K.