[论文解读] Diverse Trajectory Forecasting with Determinantal Point Processes

Ye Yuan, Kris Kitani|arXiv (Cornell University)|Jul 11, 2019

Point processes and geometric inequalities参考文献 46被引用 67

一句话总结

本文提出一种多样性采样函数（DSF），在基于DPP的多样性损失引导下，生成来自cVAE解码的未来轨迹的多样且可能集合，从而改进多模态轨迹预测。

ABSTRACT

The ability to forecast a set of likely yet diverse possible future behaviors of an agent (e.g., future trajectories of a pedestrian) is essential for safety-critical perception systems (e.g., autonomous vehicles). In particular, a set of possible future behaviors generated by the system must be diverse to account for all possible outcomes in order to take necessary safety precautions. It is not sufficient to maintain a set of the most likely future outcomes because the set may only contain perturbations of a single outcome. While generative models such as variational autoencoders (VAEs) have been shown to be a powerful tool for learning a distribution over future trajectories, randomly drawn samples from the learned implicit likelihood model may not be diverse -- the likelihood model is derived from the training data distribution and the samples will concentrate around the major mode that has most data. In this work, we propose to learn a diversity sampling function (DSF) that generates a diverse and likely set of future trajectories. The DSF maps forecasting context features to a set of latent codes which can be decoded by a generative model (e.g., VAE) into a set of diverse trajectory samples. Concretely, the process of identifying the diverse set of samples is posed as a parameter estimation of the DSF. To learn the parameters of the DSF, the diversity of the trajectory samples is evaluated by a diversity loss based on a determinantal point process (DPP). Gradient descent is performed over the DSF parameters, which in turn move the latent codes of the sample set to find an optimal diverse and likely set of trajectories. Our method is a novel application of DPPs to optimize a set of items (trajectories) in continuous space. We demonstrate the diversity of the trajectories produced by our approach on both low-dimensional 2D trajectory data and high-dimensional human motion data.

研究动机与目标

• 在安全关键感知系统中说明需要多样化的未来轨迹集合的动机。
• 提出一个DSF，将预测上下文映射到由cVAE解码的潜在编码，从而生成多样化样本。
• 利用判定点过程（DPP）的多样性损失，通过梯度下降优化DSF参数。
• 展示DSF在低维和高维数据下，相较基线能产生更具多样性且更具代表性的轨迹集合。

提出的方法

• 训练一个条件变分自编码器（cVAE）以建模未来轨迹的p(x|ψ)。
• 引入参数γ的DSF神经网络，输出用于解码成轨迹的一组潜在编码z。
• 使用Di 序列对角矩阵L = Diag(r) S Diag(r)，其中S是对轨迹的高斯样式相似性，r是潜在空间质量向量，定义基于DPP的多样性损失。
• 将多样性目标定义为L_diverse(γ) = -tr(I - (L(γ) + I)^{-1})，并通过梯度下降优化γ。
• 在推理阶段，生成一个DSF基准集，并（可选）对DPP执行MAP推断以选择一个多样的子集。
• 使用来自DSF推导的潜在编码的N个样本并用cVAE解码器解码得到Y。

实验结果

研究问题

• RQ1 学习到的DSF是否能产生比标准从cVAE采样更具多样性和代表性的未来轨迹集？
• RQ2 在平衡与不平衡数据下，以及在低维和高维轨迹预测任务中，DSF的表现如何？
• RQ3 基于DPP的多样性目标是否在训练中稳定并在不牺牲轨迹质量的前提下提升多样性？

主要发现

• 在平衡和不平衡条件下，DSF在合成数据上以多样性相关指标持续优于基线（cVAE、MCL、R2P2、cGAN）。
• 在合成二维数据（N=10）上，DSF在平衡/不平衡下的ADE分别为0.182/0.198，FDE分别为0.344/0.371，ASD与FSD较高，表示重复性较低。
• 在人类动作数据上，DSF得到更低的ADE/FDE（如N=10时0.259/0.421）且多样性更高（ASD 0.115，FSD 0.282）相较基线。
• 在大规模人体动作数据Human3.6M的实验（N=10和N=50）中，DSF变体在多样性与质量之间取得有利的权衡，DSF在ADE/FDE方面具有竞争力，在若干配置中多样性指标显著更高。