人工智能
姿势
计算机科学
计算机视觉
三维姿态估计
匹配(统计)
模式识别(心理学)
转化(遗传学)
方向(向量空间)
代表(政治)
对象(语法)
人工神经网络
图像(数学)
过程(计算)
数学
基因
统计
操作系统
政治
生物化学
化学
法学
政治学
几何学
作者
Li Yi,Wen Gu,Xiangyang Ji,Yu Xiang,Dieter Fox
标识
DOI:10.1007/s11263-019-01250-9
摘要
Estimating 6D poses of objects from images is an important problem in various applications such as robot manipulation and virtual reality. While direct regression of images to object poses has limited accuracy, matching rendered images of an object against the input image can produce accurate results. In this work, we propose a novel deep neural network for 6D pose matching named DeepIM. Given an initial pose estimation, our network is able to iteratively refine the pose by matching the rendered image against the observed image. The network is trained to predict a relative pose transformation using a disentangled representation of 3D location and 3D orientation and an iterative training process. Experiments on two commonly used benchmarks for 6D pose estimation demonstrate that DeepIM achieves large improvements over state-of-the-art methods. We furthermore show that DeepIM is able to match previously unseen objects.
科研通智能强力驱动
Strongly Powered by AbleSci AI