IEEE Trans Vis Comput Graph. 2017 Nov;23(11):2410-2418. doi: 10.1109/TVCG.2017.2734599. Epub 2017 Aug 10.
We present a temporal 6-DOF tracking method which leverages deep learning to achieve state-of-the-art performance on challenging datasets of real world capture. Our method is both more accurate and more robust to occlusions than the existing best performing approaches while maintaining real-time performance. To assess its efficacy, we evaluate our approach on several challenging RGBD sequences of real objects in a variety of conditions. Notably, we systematically evaluate robustness to occlusions through a series of sequences where the object to be tracked is increasingly occluded. Finally, our approach is purely data-driven and does not require any hand-designed features: robust tracking is automatically learned from data.
我们提出了一种基于深度学习的 6 自由度时间跟踪方法，在真实世界捕捉的具有挑战性的数据集上实现了最先进的性能。与现有的表现最佳的方法相比，我们的方法不仅更准确，而且对遮挡更鲁棒，同时保持实时性能。为了评估其效果，我们在各种条件下对几个具有挑战性的真实物体 RGBD 序列进行了评估。值得注意的是，我们通过一系列序列系统地评估了对遮挡的鲁棒性，在这些序列中，要跟踪的对象被逐渐遮挡。最后，我们的方法是纯粹的数据驱动的，不需要任何手工设计的特征：从数据中自动学习鲁棒跟踪。
