Feature Matching Approach for 3D Reconstruction from Multiple Images

This paper handles a robust method to handlematching problem of the 3D building from remotesensed images, where two or more images are capturedfrom the different views without any camerainformation. The system is based on the SIFT interestpoint detector on foreground object. The systemprovides a method for extracting distinctive invariantfeatures from images that can be used to performreliable matching between different view of an objector scene. The features are invariant to image scaling,translation, and rotation, and partially invariant toillumination changes and affine or 3D projection. Thesystem also describes the outlier removal and thehomography matrix computation approach betweentwo planes using corresponding points based on theRANSAC method. The proposed method gives asufficient number of matches distributed on the imageand is particularly suitable for 3D reconstruction fromfeature correspondence