AffectSim：用于具身情感感知的可控交互式3D仿真基准
AffectSim: A Controllable Interactive 3D Simulation Benchmark for Embodied Affective Perception
浏览论文内容
中文总结 AI 辅助
本文提出AffectSim这一可控交互式3D仿真基准，通过分离情感行为与观测条件实现可控感知，实验显示主动观测基线可提升多数感知模型性能，该基准为具身情感感知研究提供了新平台。
中文摘要 AI 辅助
现有情感基准大多由固定的录音组成，其观测条件在推理前就已确定，难以系统研究具身感知如何影响情感感知。我们提出AffectSim，这是一个用于具身情感感知的可控交互式3D仿真基准。与将情感样本视为固定录音不同，AffectSim将表达情感的人类动作实例化为可回放的3D场景，在保留底层行为和情感标签的同时，可系统地改变距离、朝向、遮挡、场景几何结构和智能体视角。该基准包含5个情感类别、57个场景中的27647个场景片段。其因子化设计将情感行为与观测条件分离，支持对同一行为的可控重观测，以及在可执行的3D环境中由智能体控制的感知。为展示此能力，我们在匹配的初始（P-Init）、参考（P-Ref）和主动获取（A-Obs）观测下实例化具身情感感知。在24个冻结感知模型配置中，P-Ref的表现显著优于P-Init，而一个简单的两阶段主动观测基线在24个配置中的21个上实现了性能提升。开源模型的平均Macro-F1从9.89%提升至11.70%，闭源模型的平均Macro-F1从22.61%提升至24.26%，分别恢复了其P-Ref与P-Init差距的32.0%和20.1%。场景片段级恢复和路径感知评估进一步在整体识别性能之外刻画了当前基线的特征。这些结果证明了使情感观测可控的价值，并确立AffectSim为通过交互式3D仿真研究具身情感感知的初始平台。
英文摘要
Existing affective benchmarks largely rely on fixed recordings, where observation conditions are predetermined before inference. Consequently, they mainly evaluate passive affect recognition while overlooking a key question for embodied agents: how to actively acquire informative affective evidence. We introduce AffectSim, a controllable interactive 3D simulation benchmark for embodied affective perception. AffectSim represents affective behaviors as replayable 3D episodes and separates the underlying behavior from how it is observed, enabling systematic control over distance, orientation, occlusion, scene geometry, and agent viewpoint. It contains 27,647 episodes across five emotion categories and 57 scenes. Across 24 frozen perception-model configurations, we show that observation quality substantially affects recognition. To examine whether active sensing can recover this gap, we establish a training-free heuristic baseline, which improves recognition in 21 of 24 settings. Building on this, we propose Evidence-Aware Observation Gate (EAOG) to adaptively assess whether additional observation is still beneficial. These results show that affective perception depends not only on the observed behavior, but also on how evidence is acquired. By making observation acquisition experimentally controllable, AffectSim provides a foundation for studying embodied affective perception in interactive 3D environments. All benchmark assets, code, and simulation environments will be released.
