🤖 AI Summary
To address the insufficient robustness of mobile RGB-D 6DoF pose estimation under depth sensor noise, this paper systematically analyzes the impact of such noise and proposes DTTDNet, a Transformer-based architecture. Methodologically, it introduces a novel geometric feature filtering module to suppress noise-induced artifacts and incorporates a Chamfer distance loss to strengthen pose supervision. Furthermore, we construct DTTD-Mobile—the first digital twin tracking dataset explicitly designed for smartphone-grade depth noise, featuring realistic noise modeling from the iPhone 14 Pro. Evaluated on DTTD-Mobile, our method achieves a maximum improvement of 60.74 points in the ADD metric over state-of-the-art approaches, establishing a new benchmark for depth-noise-robust 6DoF pose estimation on mobile devices.
📝 Abstract
Robust 6DoF pose estimation with mobile devices is the foundation for applications in robotics, augmented reality, and digital twin localization. In this paper, we extensively investigate the robustness of existing RGBD-based 6DoF pose estimation methods against varying levels of depth sensor noise. We highlight that existing 6DoF pose estimation methods suffer significant performance discrepancies due to depth measurement inaccuracies. In response to the robustness issue, we present a simple and effective transformer-based 6DoF pose estimation approach called DTTDNet, featuring a novel geometric feature filtering module and a Chamfer distance loss for training. Moreover, we advance the field of robust 6DoF pose estimation and introduce a new dataset -- Digital Twin Tracking Dataset Mobile (DTTD-Mobile), tailored for digital twin object tracking with noisy depth data from the mobile RGBD sensor suite of the Apple iPhone 14 Pro. Extensive experiments demonstrate that DTTDNet significantly outperforms state-of-the-art methods at least 4.32, up to 60.74 points in ADD metrics on the DTTD-Mobile. More importantly, our approach exhibits superior robustness to varying levels of measurement noise, setting a new benchmark for the robustness to noise measurements. Code and dataset are made publicly available at: https://github.com/augcog/DTTD2
