Method for visual search, corresponding system, apparatus and computer program product

1. A method, comprising:

extracting, using digital image processing circuitry, local descriptors from digital image information and digital depth information related to one or more digital images, the local descriptors conveying appearance description information and shape description information related to said one or more digital images, wherein the digital image information is generated using one or more RGB sensors and the digital depth information is generated using one or more depth sensors;

generating, using the digital image processing circuitry, one or more global representations of the one or more digital images based on the extracted local descriptors;

hashing, using the digital image processing circuitry, the one or more global representations of the one or more digital images; and

generating, using the digital image processing circuitry, one or more visual search queries based on the hashed one or more global representations, wherein the one or more visual search queries include fused appearance description information and shape description information conveyed in the local descriptors extracted from the digital image information generated using one or more RGB sensors and the digital depth information generated using one of more depth sensors.

2. The method of claim 1 , comprising:

fusing local descriptors extracted from said digital image information and local descriptors extracted from said digital depth information; and

generating the one or more global representations of the one or more digital images based on the fused local descriptors.

3. The method of claim 1 , comprising:

fusing appearance description information and shape description information included in the one or more global representations; and

subsequently hashing the one or more global representations.

4. The method of claim 1 , comprising fusing appearance description information and shape description information after the hashing of the one or more global representations.

5. The method of claim 1 wherein the digital image information includes digital color information.

6. The method of claim 5 wherein the one or more digital images comprise RGB-D digital images.

7. The method of claim 1 , comprising:

selectively varying respective amounts of said appearance description information and shape description information included in the fused appearance description information and shape description information of the visual search queries.

8. The method of claim 1 , comprising:

sensing the digital image information and the digital depth information of an image of the one or more digital images using an RGB-D image sensor including the one or more RGB sensors and the one or more depth sensors.

9. An apparatus, comprising:

one or more memories, which, in operation, store digital image information; and

digital image processing circuitry coupled to the one or more memories, wherein the digital image processing circuitry, in operation:

extracts local descriptors from RGB sensor digital image information and depth sensor digital depth information related to one or more digital images, the local descriptors conveying appearance description information and shape description information related to said one or more digital images;

generates one or more global representations of the one or more digital images based on the extracted local descriptors;

hashes the one or more global representations of the one or more digital images; and

generates one or more visual search queries based on the hashed one or more global representations, wherein the one or more visual search queries include fused appearance description information and shape description information conveyed in the local descriptors, the digital image information is generated using one or more RGB sensors and the digital depth information is generated using one of more depth sensors.

10. The apparatus of claim 9 wherein the digital image processing circuitry, in operation:

fuses local descriptors extracted from said digital image information and local descriptors extracted from said digital depth information; and

generates the one or more global representations of the one or more digital images based on the fused local descriptors.

11. The apparatus of claim 9 wherein the digital image processing circuitry, in operation:

fuses appearance description information and shape description information included in the one or more global representations; and

hashes the one or more global representations including the fused appearance description information and shape description information.

12. The apparatus of claim 9 wherein the digital image processing circuitry, in operation, fuses appearance description information and shape description information included in the hashed one or more global representations.

13. The apparatus of claim 9 wherein the one or more digital images comprise RGB-D digital images.

14. The apparatus of claim 9 wherein the digital image processing circuitry, in operation, selectively varies respective amounts of said appearance description information and shape description information included in the fused appearance description information and shape description information of the visual search queries.

15. The apparatus of claim 14 wherein the digital image processing circuitry, in operation, selective varies the respective amounts in response to a visual search response signal.

16. The apparatus of claim 9 , comprising an integrated circuit including the digital image processing circuitry.

17. The apparatus of claim 9 , comprising one or more RGB-D image sensors.

18. A system, comprising:

one or more image capture devices, which, in operation, capture digital images, the one or more image capture devices including one or more RGB image sensors and one or more digital depth sensors; and

digital image processing circuitry coupled to the one or more image capture devices, wherein the digital image processing circuitry, in operation:

extracts local descriptors from RGB digital image information and digital depth information related to one or more captured digital images, the local descriptors conveying appearance description information and shape description information related to said one or more digital images;

generates one or more global representations of the one or more digital images based on the extracted local descriptors;

hashes the one or more global representations of the one or more digital images; and

generates one or more visual search queries based on the hashed one or more global representations, wherein the one or more visual search queries include fused appearance description information and shape description information conveyed in the local descriptors, the digital image information is generated using the one or more RGB sensors and the digital depth information is generated using the one of more depth sensors.

19. The system of claim 18 wherein the digital image processing circuitry, in operation:

fuses local descriptors extracted from said digital image information and local descriptors extracted from said digital depth information; and

generates the one or more global representations of the one or more digital images based on the fused local descriptors.

20. The system of claim 18 wherein the digital image processing circuitry, in operation:

fuses appearance description information and shape description information included in the one or more global representations; and

hashes the one or more global representations including the fused appearance description information and shape description information.

21. The system of claim 18 wherein the digital image processing circuitry, in operation, fuses appearance description information and shape description information included in the hashed one or more global representations.

22. The system of claim 18 wherein the digital image processing circuitry, in operation, selectively varies respective amounts of said appearance description information and shape description information included in the fused appearance description information and shape description information of the visual search queries.

23. A non-transitory computer-readable medium having contents which cause digital image processing circuitry to generate one or more visual search queries, by:

extracting local descriptors from digital image information and digital depth information related to one or more digital images, the local descriptors conveying appearance description information and shape description information related to said one or more digital images, wherein the digital image information is generated using one or more RGB sensors and the digital depth information is generated using one or more depth sensors;

generating one or more global representations of the one or more digital images based on the extracted local descriptors;

hashing the one or more global representations of the one or more digital images; and

generating one or more visual search queries based on the hashed one or more global representations, wherein the one or more visual search queries include fused appearance description information and shape description information conveyed in the local descriptors extracted from the digital image information generated using one or more RGB sensors and the digital depth information generated using one of more depth sensors.

24. The non-transitory computer-readable medium of claim 23 wherein the contents cause the digital image processing circuitry to perform at least one of:

fusing local descriptors extracted from said digital image information and local descriptors extracted from said digital depth information;

fusing appearance description information and shape description information included in the one or more global representations; and

fusing appearance description information and shape description information included in the hashed one or more global representations.

25. The non-transitory computer-readable medium of claim 23 wherein the contents cause the digital image processing circuitry to selectively vary respective amounts of said appearance description information and shape description information included in the fused appearance description information and shape description information of the visual search queries.