HEIR：以功能角色学习人-实体交互
原标题：HEIR: Learning Human-Entity Interactions with Functional Roles
AI 摘要
该论文提出 HEIR 基准与 CoRISP 方法，用于从单张图像中恢复完整的人-实体交互事件结构，包括参与者、功能角色与共享实体身份。HEIR 包含 18,730 张图像、六种角色、105 种动作和 437 个名词，并引入完整集合 AP 与结构评估指标。CoRISP 在重复角色事件和共享参与者图像上分别领先 2.87 和 3.82 Set mAP，在 V-COCO 上也取得角色 AP 与完整集合 AP 结果，表明事件组合学习与评估的价值。
正文节选
HEIR: Learning Human-Entity Interactions with Functional Roles Abstract Understanding human–entity interactions requires recovering each person–action event’s participants, roles, and shared identities. This structure can support embodied agents by clarifying who acts on which entities and how, informing anticipation and coordination in shared environments. Standard HOI metrics score individual links, leaving complete event composition undermeasured. We introduce HEIR (Human–Entity Interactions