arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2302.05041cs.CVcs.LGcs.RO
收藏

基于空间对齐时间编码的图像驱动数据驱动随机运动评估与优化

Data-Driven Stochastic Motion Evaluation and Optimization with Image by Spatially-Aligned Temporal Encoding

Graduate School of Engineering, Toyota Technological Institute(丰田工业大学大学院工学研究科)

Takeru Oba, Norimichi Ukita

AI总结:

针对长时概率运动预测中跨域一致性评估难、优化易陷局部最优的问题,提出带空间对齐时间编码的EBM评估方法及自监督DMO优化器,经实验验证效果优于同类SOTA方法。

摘要:

本文提出了一种面向长时运动的概率运动预测方法,该方法可预测运动过程,使其从给定图像中观测到的初始状态出发完成指定任务。我们的方法通过基于能量的模型(Energy-Based Model,EBM)评估任务可完成性,但以往的EBM并非为评估不同域(即本文中的图像与运动)之间的一致性而设计。本方法通过空间对齐时间编码,将图像与运动数据无缝整合到图像特征域中,沿投影到图像上的运动轨迹提取特征。此外,本文还提出了一种数据驱动的运动优化方法——深度运动优化器(Deep Motion Optimizer,DMO),可与EBM协同用于运动预测。与以往基于梯度的优化器不同,我们的自监督DMO缓解了超参数调优的难度,可避免陷入局部最优。通过与同类SOTA方法开展多组实验,验证了所提方法的有效性。

英文摘要:

This paper proposes a probabilistic motion prediction method for long motions. The motion is predicted so that it accomplishes a task from the initial state observed in the given image. While our method evaluates the task achievability by the Energy-Based Model (EBM), previous EBMs are not designed for evaluating the consistency between different domains (i.e., image and motion in our method). Our method seamlessly integrates the image and motion data into the image feature domain by spatially-aligned temporal encoding so that features are extracted along the motion trajectory projected onto the image. Furthermore, this paper also proposes a data-driven motion optimization method, Deep Motion Optimizer (DMO), that works with EBM for motion prediction. Different from previous gradient-based optimizers, our self-supervised DMO alleviates the difficulty of hyper-parameter tuning to avoid local minima. The effectiveness of the proposed method is demonstrated with a variety of experiments with similar SOTA methods.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑