Transmutation of virtual entity sketch using extracted features and relationships of real and virtual objects in mixed reality scene
A method, system and computer program product for enriching user sketched virtual objects in mixed reality scenes to provide more relevant content. Objects, including a virtual entity sketch and real and virtual objects, are identified in a mixed reality scene. The features for each of these objects are extracted, such as color, texture, parts, etc. A relative position and connectedness of the virtual entity sketch are computed with respect to the mixed reality scene as an object-object matrix using the extracted features. The object-object matrix is then augmented using open and closed regions of the mixed reality scene to create an inter-object relationship graph. The virtual entity sketch is then transmuted using the inter-object relationship graph. Content may then be retrieved using the transmuted virtual entity sketch. In this manner, content that is more relevant to the user can be retrieved based on augmenting and enriching the virtual entity sketch.
1. A method for enriching user sketched virtual objects in mixed reality scenes to provide more relevant content, the method comprising:
identifying objects in a mixed reality scene, wherein said objects comprise a virtual entity sketch, one or more real objects and one or more virtual objects;
extracting features for each of said identified objects;
computing a relative position and connectedness of said virtual entity sketch with respect to said mixed reality scene as an object-object matrix using said extracted features;
augmenting said object-object matrix using open and closed regions of said mixed reality scene to create an inter-object relationship graph; and
transmuting said virtual entity sketch using said inter-object relationship graph.
2. The method as recited in claim 1 further comprising:
retrieving content using said transmuted virtual entity sketch.
3. The method as recited in claim 1 further comprising:
converting said identified real and virtual objects to sketches using an edge detector technique;
converting strokes of said sketches to shape words using a shape word descriptor;
storing said extracted features and said shape words in a data store; and
computing said relative position and connectedness of said virtual entity sketch with respect to said mixed reality scene as said object-object matrix using said extracted features and said shape words.
4. The method as recited in claim 1 , wherein said closed regions are regions containing only interior pixels, wherein said open regions are regions containing interior and border pixels.
5. The method as recited in claim 1 further comprising:
segmenting said virtual entity sketch into regions; and
clustering objects to which said regions share connectively.
6. The method as recited in claim 1 , wherein said transmutation of said virtual entity sketch involves a transmutation of one or more of the following features of said virtual entity sketch: contrast, symmetry, style, aspect ratio and viewpoint.
7. The method as recited in claim 1 , wherein said transmutation of said virtual entity sketch comprises one or more of the following: grouping, peak shift and isolation principle, visual metaphors and perceptual problem solving.
8. A computer program product for enriching user sketched virtual objects in mixed reality scenes to provide more relevant content, the computer program product comprising a computer readable storage medium having program code embodied therewith, the program code comprising the programming instructions for:
identifying objects in a mixed reality scene, wherein said objects comprise a virtual entity sketch, one or more real objects and one or more virtual objects;
extracting features for each of said identified objects;
computing a relative position and connectedness of said virtual entity sketch with respect to said mixed reality scene as an object-object matrix using said extracted features;
augmenting said object-object matrix using open and closed regions of said mixed reality scene to create an inter-object relationship graph; and
transmuting said virtual entity sketch using said inter-object relationship graph.
9. The computer program product as recited in claim 8 , wherein the program code further comprises the programming instructions for:
retrieving content using said transmuted virtual entity sketch.
10. The computer program product as recited in claim 8 , wherein the program code further comprises the programming instructions for:
converting said identified real and virtual objects to sketches using an edge detector technique;
converting strokes of said sketches to shape words using a shape word descriptor;
storing said extracted features and said shape words in a data store; and
computing said relative position and connectedness of said virtual entity sketch with respect to said mixed reality scene as said object-object matrix using said extracted features and said shape words.
11. The computer program product as recited in claim 8 , wherein said closed regions are regions containing only interior pixels, wherein said open regions are regions containing interior and border pixels.
12. The computer program product as recited in claim 8 , wherein the program code further comprises the programming instructions for:
segmenting said virtual entity sketch into regions; and
clustering objects to which said regions share connectively.
13. The computer program product as recited in claim 8 , wherein said transmutation of said virtual entity sketch involves a transmutation of one or more of the following features of said virtual entity sketch: contrast, symmetry, style, aspect ratio and viewpoint.
14. The computer program product as recited in claim 8 , wherein said transmutation of said virtual entity sketch comprises one or more of the following: grouping, peak shift and isolation principle, visual metaphors and perceptual problem solving.
15. A system, comprising:
a memory for storing a computer program for enriching user sketched virtual objects in mixed reality scenes to provide more relevant content; and
a processor connected to said memory, wherein said processor is configured to execute the program instructions of the computer program comprising:
identifying objects in a mixed reality scene, wherein said objects comprise a virtual entity sketch, one or more real objects and one or more virtual objects;
extracting features for each of said identified objects;
computing a relative position and connectedness of said virtual entity sketch with respect to said mixed reality scene as an object-object matrix using said extracted features;
augmenting said object-object matrix using open and closed regions of said mixed reality scene to create an inter-object relationship graph; and
transmuting said virtual entity sketch using said inter-object relationship graph.
16. The system as recited in claim 15 , wherein the program instructions of the computer program further comprise:
retrieving content using said transmuted virtual entity sketch.
17. The system as recited in claim 15 , wherein the program instructions of the computer program further comprise:
converting said identified real and virtual objects to sketches using an edge detector technique;
converting strokes of said sketches to shape words using a shape word descriptor;
storing said extracted features and said shape words in a data store; and
computing said relative position and connectedness of said virtual entity sketch with respect to said mixed reality scene as said object-object matrix using said extracted features and said shape words.
18. The system as recited in claim 15 , wherein said closed regions are regions containing only interior pixels, wherein said open regions are regions containing interior and border pixels.
19. The system as recited in claim 15 , wherein the program instructions of the computer program further comprise:
segmenting said virtual entity sketch into regions; and
clustering objects to which said regions share connectively.
20. The system as recited in claim 15 , wherein said transmutation of said virtual entity sketch involves a transmutation of one or more of the following features of said virtual entity sketch: contrast, symmetry, style, aspect ratio and viewpoint.