LidaRefer：面向自动驾驶的上下文感知户外3D视觉定位
LidaRefer: Context-aware Outdoor 3D Visual Grounding for Autonomous Driving
AI总结：
针对户外3D视觉定位面临的背景干扰大、上下文标注缺失问题，提出上下文感知框架LidaRefer及DiSCo监督策略与伪标注方法，在Talk2Car-3D数据集上实现最优性能。
摘要：
3D视觉定位（VG）旨在通过自然语言描述在3D场景中定位目标对象或区域。尽管室内3D视觉定位已取得进展，但户外3D视觉定位仍因两大挑战未得到充分研究：（1）大规模户外LiDAR场景以背景点为主，前景信息有限，使得跨模态对齐和上下文理解更具难度；（2）多数户外数据集缺乏指代性非目标对象的空间标注，阻碍了指代上下文的显式学习。为此，我们提出LidaRefer——一种适用户外场景的上下文感知3D视觉定位框架。LidaRefer采用以对象为中心的特征选择策略，聚焦语义相关的视觉特征，同时降低计算开销。其次，其基于Transformer的编码器-解码器架构擅长在优化后的视觉特征与词级文本特征间建立细粒度跨模态对齐，并捕捉全面的全局上下文。此外，我们提出判别-支持协同定位（DiSCo）这一新型监督策略，显式建模目标对象、上下文对象与歧义对象间的空间关系，以实现精准目标识别。为无需人工标注即可实现该策略，我们引入一种伪标注方法，可为指代性非目标对象检索3D定位标签。LidaRefer在Talk2Car-3D数据集的多种评估设置下均取得了当前最优性能。
英文摘要：
3D visual grounding (VG) aims to locate objects or regions within 3D scenes guided by natural language descriptions. While indoor 3D VG has advanced, outdoor 3D VG remains underexplored due to two challenges: (1) large-scale outdoor LiDAR scenes are dominated by background points and contain limited foreground information, making cross-modal alignment and contextual understanding more difficult; and (2) most outdoor datasets lack spatial annotations for referential non-target objects, which hinders explicit learning of referential context. To this end, we propose LidaRefer, a context-aware 3D VG framework for outdoor scenes. LidaRefer incorporates an object-centric feature selection strategy to focus on semantically relevant visual features while reducing computational overhead. Then, its transformer-based encoder-decoder architecture excels at establishing fine-grained cross-modal alignment between refined visual features and word-level text features, and capturing comprehensive global context. Additionally, we present Discriminative-Supportive Collaborative localization (DiSCo), a novel supervision strategy that explicitly models spatial relationships between target, contextual, and ambiguous objects for accurate target identification. To enable this without manual labeling, we introduce a pseudo-labeling approach that retrieves 3D localization labels for referential non-target objects. LidaRefer achieves state-of-the-art performance on Talk2Car-3D dataset under various evaluation settings.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
