发表机构
University of Toronto; Qualcomm Technologies, Inc.（多伦多大学; 高通技术公司）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
研究针对自动驾驶中物体易被遮挡问题，提出超越视野框架，通过维持持久物体假设解耦物体存在与可观测性，引入nuScenes - Permanence用于训练评估，实验证明该框架显著提升遮挡推理能力，凸显物体持久性对自动驾驶的重要性。
AI 中文摘要
自动驾驶运行在部分可观测环境中，物体可能被其他车辆或基础设施完全遮挡。大多数端到端驾驶系统将物体的存在与即时观测隐含地联系在一起，导致在长时间遮挡期间物体假设退化或消失，使潜在关键物体从下游预测和规划中消失。我们引入了超越视野（BeyondSight），这是一个具有持久性感知的端到端驾驶框架，通过随时间维持持久的物体假设，将物体的存在与可观测性解耦。超越视野在时间上传播物体查询，并用观测条件证据更新它们，即使物体暂时不可观测，也能使联合感知、预测和规划对物体进行推理。为了对具有持久性感知的模型进行有原则的训练和评估，我们进一步引入了nuScenes - Permanence，它是nuScenes的扩展，为不可观测物体提供监督和基于可观测性的评估。实验表明，超越视野显著提高了遮挡情况下的推理能力，将不可观测物体的检测性能从0提高到0.249 mAP，同时将规划误差从0.61降低到0.54 L2avg。这些结果凸显了物体持久性作为稳健端到端自动驾驶重要建模原则的地位。
英文摘要
Autonomous driving operates in partially observable environments where actors may become fully occluded by other vehicles or infrastructure. Most end-to-end driving systems implicitly couple actor existence to instantaneous observations, causing actor hypotheses to degrade or disappear during prolonged occlusion and removing potentially critical agents from downstream prediction and planning. We introduce BeyondSight, a permanence-aware end-to-end driving framework that decouples actor existence from observability by maintaining persistent actor hypotheses over time. BeyondSight propagates actor queries temporally and updates them with observation-conditioned evidence, enabling joint perception, prediction, and planning to reason about actors even when they are temporarily unobservable. To enable principled training and evaluation of persistence-aware models, we further introduce nuScenes-Permanence, an extension of nuScenes that provides supervision and observability-conditioned evaluation for unobservable actors. Experiments show that BeyondSight substantially improves reasoning under occlusion, increasing detection performance for unobservable actors from 0 to 0.249 mAP while reducing planning error from 0.61 to 0.54 L2avg. These results highlight object permanence as an important modeling principle for robust end-to-end autonomous driving.
CommentsAccepted to ECCV 2026
