Abstract
We propose a novel method for joint estimation of shape and pose of rigid objects from their sequentially observed RGB-D images. In sharp contrast to past approaches that rely on complex non-linear optimization, we propose to formulate it as a neural optimization that learns to efficiently estimate the shape and pose. We introduce Deep Directional Distance Function (DeepDDF), a neural network that directly outputs the depth image of an object given the camera viewpoint and viewing direction, for efficient error computation in 2D image space. We formulate the joint estimation itself as a Transformer which we refer to as TransPoser. We fully leverage the tokenization and multi-head attention to sequentially process the growing set of observations and to efficiently update the shape and pose with a learned momentum, respectively. Experimental results on synthetic and real data show that DeepDDF achieves high accuracy as a category-level object shape representation and TransPoser achieves state-of-the-art accuracy efficiently for joint shape and pose estimation.
Abstract (translated)
我们提出了一种新方法，用于从Sequentially observed RGB-D图像中 joint estimation rigid 对象的形状和姿态。与过去依赖于复杂的非线性优化的方法相反，我们提议将其写成一种神经网络优化，学会高效地估计形状和姿态。我们引入了 Deep Directional Distance Function (DeepDDF)，它是一个神经网络，根据相机视角和观测方向，直接输出对象的深度图像，用于在 2D 图像空间中高效计算错误。我们将其整形成一种Transformer，我们称之为TransPoser。我们 fully 利用分块和多目注意力，Sequentially 处理不断增加的观测值，并使用学习的动力更新形状和姿态。模拟数据和实际数据的实验结果显示，DeepDDF 作为类级别的对象形状表示，具有极高的准确性，而TransPoser 高效地用于 joint 形状和姿态估计。
URL
https://arxiv.org/abs/2303.13477
