Abstract: We propose a novel shape descriptor named geodesic spin contour. Geodesic spin contour is similar to the classical spin image, but it replaces spin coordinates with quantities invariant under isometric transformations: geodesic distance and average geodesic distance. This makes it applicable to matching near-isometric shapes. By choosing a small support length, one can obtain geodesic spin contours that describe the local geometry of the shape, and therefore are suitable for surfaces with missing data. Experimental results show that the geodesic spin contour provides an effective way to match features on near isometric shapes.