[1] [1] WuC. Towardslinear-timeinci emental stru Uurefrommotion［C］. 3DTV-Conference，2013 International Conference on.IEEE，2013: 127-134.

[2] [2] Huber，PeterJ. Robuststatistics［M］. JohnWiley&Sons，2004.

[3] [3] Engel J，Schops T，Cremers D. LSD-SLAM: Large-scaledirect monocular S-LAM［C］. European Conference on ComputerVision. Springer，2014: 834-849.

[4] [4] Civera J，DavisonAJ，Montiel J M. Inverse depth parametrization formonocular SLAM［J］. IEEE transactions on robotics，2008，24(5): 932-945.

[5] [5] Trajkovi.M，HedleyM. Fastcornerdetection［J］. ImageandVisionComputing，1998，16(2): 75-87.

[6] [6] MaybankSJ，FaugerasOD. Atheoryofself-calibrationof amovingcamera［J］. InternationalJournalofComputer

[7] [7] Dehais J，Anthimopoulos M，Shevchik S. Two-view 3D reconstruction for food volume estimation［J］. IEEE TransactionsonMultimedia，2016，19(5): 1090-1099.

[8] [8] Soheilian B，Paparoditis N，Vallet B. Detection and 3D reconstruction of traffic signs from multiple view color images［J］. ISPRSJournalofPhotogrammetryandRemoteSensing，2013，77(5): 1-20.

[9] [9] Shen S. Accurate multiple view 3d reconstruction using patch-based stereo for large-scale scenes［C］. IEEE TransactionsonImageProcessing，2013，22(5): 1901-1914.

[10] [10] Vincent Lepetit，Francesc Moreno-Noguer，PascalFua. EPnP: AnAccurate O(n)Solution to the Pn PProblem［J］. InternationalJournalofComputerVision，2009，(2): 33-36.

[11] [11] Alex Krizhevsky，Ilya Sutskever，Geoffrey E Hinton. ImageNet classification with deep convolutional neural networks［J］.CommunicationsoftheACM，2017，(6): 17-25.

[12] [12] HainanCui，ShuhanShen，ZhanyiHu. Globalfusionofgeneralizedcameramodelforefficientlarge-scalestructure frommotion［J］. ScienceChinaInformationSciences，2017，(3): 158-163.