TY - GEN
T1 - LOD-InfiniTAM
T2 - 7th International Conference on Robotics and Computer Vision, ICRCV 2025
AU - Zhang, Tiantian
AU - Li, Ni
AU - Gong, Guanghong
AU - Tian, Bo
AU - Lin, Xin
N1 - Publisher Copyright:
© 2025 IEEE.
PY - 2025
Y1 - 2025
N2 - Dense RGB-D Simultaneous Localization and Mapping (SLAM) is the mainstream approach for augmented reality and autonomous robotics, fields that demand high accuracy and real-time performance in markerless environments. However, the measurement noise of depth sensors increases quadratically with distance, introducing significant interference into SLAM systems. In this paper, we propose an adaptive depth sampling method based on a Level-of-Detail (LoD) algorithm that selectively filters observations from distant or textureless regions. Integrated into the InfiniTAM framework, the method reduces both noise-induced errors and redundant computation. Evaluations on TUM RGB-D dataset demonstrate that the method improves tracking accuracy by 22%, reduces artifacts in volumetric reconstruction, and maintains real-time performance at over 240 FpS.
AB - Dense RGB-D Simultaneous Localization and Mapping (SLAM) is the mainstream approach for augmented reality and autonomous robotics, fields that demand high accuracy and real-time performance in markerless environments. However, the measurement noise of depth sensors increases quadratically with distance, introducing significant interference into SLAM systems. In this paper, we propose an adaptive depth sampling method based on a Level-of-Detail (LoD) algorithm that selectively filters observations from distant or textureless regions. Integrated into the InfiniTAM framework, the method reduces both noise-induced errors and redundant computation. Evaluations on TUM RGB-D dataset demonstrate that the method improves tracking accuracy by 22%, reduces artifacts in volumetric reconstruction, and maintains real-time performance at over 240 FpS.
KW - Level-of-detail
KW - real-time volumetric reconstruction
KW - RGB-D SLAM
UR - https://www.scopus.com/pages/publications/105033041298
U2 - 10.1109/ICRCV67407.2025.11349343
DO - 10.1109/ICRCV67407.2025.11349343
M3 - 会议稿件
AN - SCOPUS:105033041298
T3 - 2025 7th International Conference on Robotics and Computer Vision, ICRCV 2025
SP - 243
EP - 247
BT - 2025 7th International Conference on Robotics and Computer Vision, ICRCV 2025
PB - Institute of Electrical and Electronics Engineers Inc.
Y2 - 24 October 2025 through 26 October 2025
ER -