Learning from the Unseen: Generative Data Augmentation for Geometric-Semantic Accident Anticipation
Learning from the Unseen: Generative Data Augmentation for Geometric-Semantic Accident Anticipation
从未见中学习：用于几何语义事故预测的生成式数据增强
Abstract: Anticipating traffic accidents is a critical yet unresolved problem for autonomous driving, hindered by the inherent complexity of modeling interactions between road users and the limited availability of diverse, large-scale datasets. 摘要： 预测交通事故是自动驾驶领域一个关键但尚未解决的问题，其难点在于道路使用者之间交互建模的固有复杂性，以及缺乏多样化的大规模数据集。
To address these issues, we propose a dual-path framework. On the one hand, we employ a video synthesis pipeline that, guided by structured prompts, derives feature distributions from existing corpora and produces high-fidelity synthetic driving scenes consistent with the statistical patterns of real data. 为了解决这些问题，我们提出了一个双路径框架。一方面，我们采用了一种视频合成流水线，在结构化提示的引导下，从现有语料库中推导出特征分布，并生成与真实数据统计模式一致的高保真合成驾驶场景。
On the other hand, we design a graph neural network enriched with semantic cues, enabling dynamic reasoning over both spatial and semantic relations among participants. 另一方面，我们设计了一种富含语义线索的图神经网络，能够对参与者之间的空间和语义关系进行动态推理。
To validate the effectiveness of our approach, we release a new benchmark dataset containing standardized, finely annotated video sequences that cover a broad spectrum of regions, weather, and traffic conditions. 为了验证我们方法的有效性，我们发布了一个新的基准数据集，其中包含标准化、精细标注的视频序列，涵盖了广泛的地区、天气和交通状况。
Evaluations across existing datasets and our new benchmark confirm notable gains in both accuracy and anticipation lead time, highlighting the capacity of the proposed framework to mitigate current data bottlenecks and enhance the reliability of autonomous driving systems. 在现有数据集和我们新基准上的评估证实，该方法在准确性和预测提前时间方面均有显著提升，凸显了所提框架在缓解当前数据瓶颈和增强自动驾驶系统可靠性方面的潜力。
Paper Details:
- Authors: Yanchen Guan, Haicheng Liao, Chengyue Wang, Xingcheng Liu, Jiaxun Zhang, Keqiang Li, Zhenning Li
- arXiv ID: 2605.00051
- Subjects: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
论文详情：
- 作者： Yanchen Guan, Haicheng Liao, Chengyue Wang, Xingcheng Liu, Jiaxun Zhang, Keqiang Li, Zhenning Li
- arXiv ID: 2605.00051
- 学科分类： 计算机视觉与模式识别 (cs.CV)；机器学习 (cs.LG)