se(3)-TrackNet: Data-driven 6D Pose Tracking by Calibrating Image Residuals in Synthetic Domains
se(3)-TrackNet: Data-driven 6D Pose Tracking by Calibrating Image Residuals in Synthetic Domains
复制标题
DOI：
10.1109/iros45743.2020.9341314
复制
发表时间：
2020-07
期刊：
影响因子：
--
通讯作者：
Bowen Wen;Chaitanya Mitash;Baozhang Ren;Kostas E. Bekris
中科院分区：
文献类型：
--
作者：
Bowen Wen;Chaitanya Mitash;Baozhang Ren;Kostas E. Bekris
跟踪视频序列中物体的6D姿态对机器人的操作非常重要。然而，这项任务引入了多重挑战：(i)机器人操作涉及明显的闭塞；（ii） 6D姿态的数据和注释很难收集，这使机器学习解决方案变得复杂；（iii）在长期跟踪中，增量误差漂移经常累积，需要重新初始化对象的姿态。这项工作提出了一种数据驱动的长期6D姿态跟踪优化方法。它的目的是在给定当前RGB-D观测和基于先前最佳估计和对象模型的合成图像的情况下识别最佳相对姿态。在这种情况下的关键贡献是一种新的神经网络架构，它适当地解开特征编码以帮助减少域移位，并通过李代数有效地表示三维方向。因此，即使网络只使用合成数据进行训练，也可以有效地处理真实图像。在基准测试上的综合实验——现有的以及与对象操作相关的显著遮挡的新数据集——表明，所提出的方法实现了一致的鲁棒估计，并且优于其他方法，即使它们已经用真实图像进行了训练。该方法也是所有备选方案中计算效率最高的，并实现了90.9Hz的跟踪频率。1
Tracking the 6D pose of objects in video sequences is important for robot manipulation. This task, however, introduces multiple challenges: (i) robot manipulation involves significant occlusions; (ii) data and annotations are troublesome and difficult to collect for 6D poses, which complicates machine learning solutions, and (iii) incremental error drift often accumulates in long term tracking to necessitate re-initialization of the object’s pose. This work proposes a data-driven optimization approach for long-term, 6D pose tracking. It aims to identify the optimal relative pose given the current RGB-D observation and a synthetic image conditioned on the previous best estimate and the object’s model. The key contribution in this context is a novel neural network architecture, which appropriately disentangles the feature encoding to help reduce domain shift, and an effective 3D orientation representation via Lie Algebra. Consequently, even when the network is trained only with synthetic data can work effectively over real images. Comprehensive experiments over benchmarks - existing ones as well as a new dataset with significant occlusions related to object manipulation - show that the proposed approach achieves consistently robust estimates and outperforms alternatives, even though they have been trained with real images. The approach is also the most computationally efficient among the alternatives and achieves a tracking frequency of 90.9Hz. 1