Wu Xiuen, Li Sien, Wang Tao, Xu Ge, Papageorgiou George
Fujian Provincial Key Laboratory of Information Processing and Intelligent Control, School of Computer and Big Data, Minjiang University, Fuzhou 350108, China.
College of Computer and Data Science, Fuzhou University, Fuzhou 350108, China.
Biomimetics (Basel). 2024 Jul 31;9(8):462. doi: 10.3390/biomimetics9080462.
We propose a memory-enhanced multi-stage goal-driven network (ME-MGNet) for egocentric trajectory prediction in dynamic scenes. Our key idea is to build a scene layout memory inspired by human perception in order to transfer knowledge from prior experiences to the current scenario in a top-down manner. Specifically, given a test scene, we first perform scene-level matching based on our scene layout memory to retrieve trajectories from visually similar scenes in the training data. This is followed by trajectory-level matching and memory filtering to obtain a set of goal features. In addition, a multi-stage goal generator takes these goal features and uses a backward decoder to produce several stage goals. Finally, we integrate the above steps into a conditional autoencoder and a forward decoder to produce trajectory prediction results. Experiments on three public datasets, JAAD, PIE, and KITTI, and a new egocentric trajectory prediction dataset, Fuzhou DashCam (FZDC), validate the efficacy of the proposed method.
我们提出了一种用于动态场景中自我中心轨迹预测的记忆增强多阶段目标驱动网络（ME-MGNet）。我们的关键思想是构建一个受人类感知启发的场景布局记忆，以便以自上而下的方式将先前经验中的知识转移到当前场景。具体来说，给定一个测试场景，我们首先基于场景布局记忆进行场景级匹配，以从训练数据中视觉上相似的场景中检索轨迹。接下来是轨迹级匹配和记忆过滤，以获得一组目标特征。此外，一个多阶段目标生成器获取这些目标特征，并使用反向解码器生成几个阶段目标。最后，我们将上述步骤集成到一个条件自动编码器和一个前向解码器中，以产生轨迹预测结果。在三个公共数据集JAAD、PIE和KITTI以及一个新的自我中心轨迹预测数据集福州行车记录仪（FZDC）上进行的实验验证了所提方法的有效性。
