A highly robust automatic 3D reconstruction system based on integrated optimization by point line features

Hou, Junyi; Yu, Lei; Fei, Shumin

doi:10.1016/j.engappai.2020.103879

Cited by 10 publications

(1 citation statement)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Simultaneous localization and mapping (SLAM) is one of the underlying core technologies in intelligent robots, autonomous vehicles, AR, MR and other fields [1][2][3]. At present, the most well-known and widely used traditional visual SLAM is the ORB-SLAM series and the VINS series.…”

Section: Introductionmentioning

confidence: 99%

Realtime 3D reconstruction algorithm based on improved AKAZE and multi-layer feature detection network

Pan,

Hou,

et al. 2024

Meas. Sci. Technol.

Self Cite

View full text Add to dashboard Cite

Simultaneous localization and mapping(SLAM) technology aims to solve the problem of how to determine the trajectory of robots by observing the environment and construct sparse spatial models. And the 3D reconstruction technology can further build dense spatial models based on the pose information. This paper proposes a dense 3D reconstruction algorithm based on improved accelerated-KAZE(AKAZE) features and a multi-layer feature detection network. A multi-layer feature detection neural network is constructed to replace the BoW model and optimize the loopback results, which improves the effect of pose estimation. The optimized algorithm improves the mapping effect and construct clear and robust indoor complex models. The pose estimation results show that the optimized algorithm improves the accuracy by 30% compared with ORB-SLAM2. Compared with the mapping effect of Kintinuous, ElasticFusion, ORB-SLAM2 and BundleFusion, the optimized models own the advantages of clearer details, richer information and no split layer.

show abstract

Section: Introductionmentioning

confidence: 99%