Methods, systems, and data structures for performing searches on three dimensional objects

• US 9,348,877 B2
• Filed: 04/22/2013
• Issued: 05/24/2016
• Est. Priority Date: 01/25/2003
• Status: Active Grant

• Current Assignee: Purdue Research Foundation (The Trustees of Purdue University (d/b/a Purdue University))
• Original Assignee: Purdue Research Foundation (The Trustees of Purdue University (d/b/a Purdue University))
• Inventors: Ramani, Karthik, Iyer, Natraj, Lou, Kuiyang, Jayanti, Subramaniam
• Primary Examiner(s): Chau, Dung K
• Application Number:
  US13/867,503
• Publication Number:
  US 20130311450A1
• Time in Patent Office: 1,128 Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/00 Information retrieval; Database structures therefor; File system structures therefor
• CPC Class Codes: G06F 16/248 Presentation of query results
• CPC Class Codes: G06F 16/5854 using shape and object relationship
• CPC Class Codes: G06F 2111/02 CAD in a network environment, e.g. collaborative CAD or distributed simulation
• CPC Class Codes: G06F 2111/20 Configuration CAD, e.g. designing by assembling or positioning modules selected from libraries of predesigned modules
• CPC Class Codes: G06F 2119/20 Design reuse, reusability analysis or reusability optimisation
• CPC Class Codes: G06F 30/10 Geometric CAD
• CPC Class Codes: G06F 30/27 using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
• CPC Class Codes: G06V 10/34 Smoothing or thinning of the pattern; Morphological operations; Skeletonisation
• CPC Class Codes: G06V 20/64 Three-dimensional objects
• CPC Class Codes: G06V 30/1988 Graph matching graphical image representation G06V30/18181

Abstract

Techniques are provided for searching on three dimensional (3D) objects across large, distributed repositories of 3D models. 3D shapes are created for input to a search system; optionally user-defined similarity criterion is used, and search results are interactively navigated and feedback received for modifying the accuracy of the search results. Search input can also be given by picking 3D models from a cluster map or by providing the orthographic views for the 3D model. Feedback can be given by a searcher as to which models are similar and which are not. Various techniques adjust the search results according to the feedback given by the searcher and present the new search results to the searcher.

Claims

1. A method of presenting search results, comprising:

presenting three dimensional objects within one or more related clusters as an answer set in response to a first search query;

interactively receiving, from a user, weighted relevance information associated with selective ones of the three dimensional objects or selective ones of the related clusters, and wherein the user classifies the selective ones of the three dimensional objects or the selective ones of the related clusters into groups identified by the user as being relevant and as being irrelevant, the relevant and irrelevant groups at least partially representing the weighted relevance information; and

interactively issuing one or more second search queries based on the received relevance information to remove a number of the three dimensional objects from the answer set and to reduce a number of the related clusters from the answer set.

2. The method of claim 1 wherein the presenting further includes presenting the three dimensional objects within the one or more related clusters as answer set in response to the first query where the first query is a three dimensional search request derived from an initial two dimensional search request.

3. The method of claim 1 wherein the receiving of the relevance information further includes receiving selective feature vectors as a portion of the relevance information, wherein the selective feature vectors define selective ones of the three dimensional objects in more than one dimension.

4. The method of claim 3 wherein the receiving of the relevance information further includes defining the selective feature vectors based on lower level constructs associated with the selective ones of the three dimensional objects.

5. The method of claim 1 wherein the presenting further includes presenting the answer set in response to the first search query where the first search query is obtained by interactively traversing a hierarchy of available three dimensional models and selecting one of the available three dimension modes as the first search query.

6. The method of claim 1 wherein the presenting further includes forming the one or more related clusters using a Self Organizing Map (SOM) clustering technique.

7. The method of claim 6 wherein the presenting further includes using a SOM where feature vectors for each of the related clusters include a combination of a cluster feature vector for the three dimensional objects included in each cluster and graph distance measures obtained from comparisons of components associated with each of the three dimensional models included in each cluster.

8. The method of claim 1 comprising the further step of searching one or more data stores with a three dimensional representation as the first search query.

9. The method of claim 8 comprising the further step of converting the three dimensional representation into a graph skeleton defining a graph data structure, wherein the graph data structure is the first search query.

10. The method of claim 8 comprising the further steps of:

enabling the three dimensional representation to be sketched; and

receiving the sketch.

11. The method of claim 1 comprising the further step of searching one or more data stores with a three dimensional representation as one of the second search queries.

12. The method of claim 1 comprising the further step of deploying a neural network to use the relevance information to identify a pattern in relevance feedback from the use, wherein the step of interactively issuing one or more second search queries is dependent upon the identified pattern.