论文标题
运动样式转移：模块化低级适应以进行深度运动预测
Motion Style Transfer: Modular Low-Rank Adaptation for Deep Motion Forecasting
论文作者
论文摘要
经过大量数据训练，深度运动预测模型取得了巨大的成功。但是，当培训数据受到限制时，它们的表现通常很差。为了应对这一挑战，我们提出了一种转移学习方法，以有效地将预先训练的预测模型适应新领域，例如看不见的代理类型和场景环境。与更新整个编码器的常规微调方法不同，我们的主要思想是减少可以精确解释目标域特异性运动样式的可调参数的量。为此，我们介绍了两个组件来利用我们先前的运动风格知识转移：（i）一个低级运动样式适配器，该适配器在低维瓶颈上投射和调整样式功能； （ii）一种模块化适配器策略，它可以解散场景上下文和运动历史的特征，以促进适应层的细粒度选择。通过广泛的实验，我们表明我们提出的适配器设计（创造的MOSA）优于几种预测基准的先验方法。
Deep motion forecasting models have achieved great success when trained on a massive amount of data. Yet, they often perform poorly when training data is limited. To address this challenge, we propose a transfer learning approach for efficiently adapting pre-trained forecasting models to new domains, such as unseen agent types and scene contexts. Unlike the conventional fine-tuning approach that updates the whole encoder, our main idea is to reduce the amount of tunable parameters that can precisely account for the target domain-specific motion style. To this end, we introduce two components that exploit our prior knowledge of motion style shifts: (i) a low-rank motion style adapter that projects and adjusts the style features at a low-dimensional bottleneck; and (ii) a modular adapter strategy that disentangles the features of scene context and motion history to facilitate a fine-grained choice of adaptation layers. Through extensive experimentation, we show that our proposed adapter design, coined MoSA, outperforms prior methods on several forecasting benchmarks.
