Home /Research /A 6D-ICP approach for 3D reconstruction and motion estimate of unknown and non-cooperative target
OTHER

A 6D-ICP approach for 3D reconstruction and motion estimate of unknown and non-cooperative target

Ke Wang, Hong Liu, Bensheng Guo, Yunfeng Gao

Year
2016
Citations
10

Abstract

For on-orbit maintenance and service of space robot, this paper presents a 6D-ICP approach for 3D reconstruction and state estimate of unknown non-cooperative target based on stereo vision. Optical flow is applied to target image features tracking in each monocular camera extracted by Harris detector across successive frames. The features tracked successively in each camera will be matched between two cameras, and corresponding 3D structure points on the targets are generated accordingly. We use 6D ICP-based method to estimate the target state and reconstruct 3D target gradually. Final experimental results are shown to demonstrate the feasibility of the proposed method.

Keywords

Computer visionArtificial intelligenceComputer scienceTracking (education)MonocularOptical flow3D reconstructionDetectorStereopsisIterative reconstruction

Related papers

Browse all OTHER papers