Ameesh Makadia - New York NY,
Brian Brewington - Fort Collins CO,
Mark Limber - Boulder CO,
Corey Michael Joshua Goldfeder - New York NY,
Walter Thatcher Ulrich - New York NY,
Google Inc. - Mountain View CA
707723, 707706, 707748, 707758, 707769, 707770, 715757, 715848, 715850, 715852, 715968, 345653, 345664, 345679
Systems and methods for 3D shape retrieval are described herein. An embodiment includes obtaining a query 3D model, determining a similarity between the query 3D model and one or more 3D models based on a plurality of features of the query 3D model and corresponding features of the one or more 3D models, retrieving one or more similar 3D models based on the determining step and ranking the one or more similar 3D models based on the similarity. The embodiment further includes computing a feature distance between the query 3D model and the one or more 3D models for each feature, from a first set of features, associated with the query 3D model and the one or more 3D models and evaluating a first composite distance function using each feature distance. Another embodiment includes constructing a retrieval structure using each of the one or more 3D models for each feature associated with the one or more 3D models and searching the retrieval structure, for each feature, to determine 3D models similar to the query 3D model.