TR2015-051

Analysis of Shape Assumptions in 3D Reconstruction of Retina from Multiple Fundus Images

    •  Ataer-Cansizoglu, E., Taguchi, Y., Kalpathy-Cramer, J., Chiang, M.F., Erdogmus, D., "Analysis of Shape Assumptions in 3D Reconstruction of Retina from Multiple Fundus Images", IEEE International Symposium on Biomedical Imaging (ISBI), DOI: 10.1109/​ISBI.2015.7164162, April 2015, pp. 1502-1505.
      BibTeX TR2015-051 PDF
      • @inproceedings{Ataer-Cansizoglu2015apr,
      • author = {Ataer-Cansizoglu, E. and Taguchi, Y. and Kalpathy-Cramer, J. and Chiang, M.F. and Erdogmus, D.},
      • title = {{Analysis of Shape Assumptions in 3D Reconstruction of Retina from Multiple Fundus Images}},
      • booktitle = {IEEE International Symposium on Biomedical Imaging (ISBI)},
      • year = 2015,
      • pages = {1502--1505},
      • month = apr,
      • publisher = {IEEE},
      • doi = {10.1109/ISBI.2015.7164162},
      • url = {https://www.merl.com/publications/TR2015-051}
      • }
  • Research Area:

    Computer Vision

Abstract:

Utilizing priors about the shape of retinal surface is important for accurate reconstruction. We present a detailed analysis of geometrical shape priors in the 3D reconstruction of retina. We first approximate the retinal surface either as a sphere inspired by the actual shape of the eyeball, or as a plane inspired by the 2D mosaicing approaches. Based on this approximation, we perform an initial camera localization with a 2D-to- 3D registration procedure. Then, parameters of the surface and the camera poses are refined through a nonlinear least squares optimization using different shape priors. The resulting 3D model and camera poses can be used for intuitively visualizing the retinal images with a model-guided browsing interface.