We present a framework for joint analysis of class-labeled uncurated 2D image and 3D model collections. Using this framework we show how to significantly improve 3D retrieval by exploiting the...

We propose a joint embedding space populated by both shapes and images, where the Euclidean distance between embedded entities reflects the similarity between the objects contained or represented by...

We present shape edit distance (SHED), a similarity measure that estimates the amount of effort needed to transform one shape into the other, in terms of rearranging the parts of one shape to match...

We present a deformation-driven approach to topology-varying 3D shape correspondence. Our method produce a fine-grained correspondence between two 3D shapes that differ in their geometry and...

A fast and one-click method produces a coarse quad layout over the triangle mesh of a digital shape using its curve skeleton. The user can steer and refine the process through interactive tools