SKETCH: Semantic Key-Point Conditioning for Long-Horizon Vessel Trajectory Prediction
SKETCH: 面向长时域船舶轨迹预测的语义关键点条件建模
机构 * School of Data Science, The Chinese University of Hong Kong, Shenzhen, China(香港中文大学(深圳)数据科学学院) ; School of Science and Engineering, The Chinese University of Hong Kong, Shenzhen, China(香港中文大学(深圳)科学与工程学院) ; School of Artificial Intelligence, The Chinese University of Hong Kong, Shenzhen, China(香港中文大学(深圳)人工智能学院) ; COSCO SHIPPING Advanced Technology Institute, Shanghai, China(中远海运技术研究院)
AI总结 针对长时域轨迹预测中方向漂移问题,提出基于语义关键点(NKP)的条件轨迹建模框架,将预测分解为全局语义决策与局部运动建模,采用预训练-微调策略估计NKP先验,在真实AIS数据上显著提升长时域、方向精度和细粒度预测性能。
Comments Final Camera-Ready Version. Accepted by ICML 2026 (PMLR Vol. 306)