Object based image based search

1. A computing system, comprising:

at least one processor;

memory including instructions that, when executed by the at least one processor, cause the computing system to at least:

process a first image that includes a plurality of objects to detect each of the plurality of objects;

for each of the plurality of objects, associate a token with the object;

present the first image that includes the plurality of objects;

receive a first selection of a first object of the plurality of objects from the first image;

receive a second selection of a second object of the plurality of objects from the first image;

determine a first token corresponding to the first object and a second token corresponding to the second object;

query at least one index maintained in a data store to determine, based at least in part on the first token or the second token, a second image indicated in the at least one index, the second image including a third object having a first visual similarity to the first object and does not include an object having a second visual similarity to the second object, wherein the first visual similarity is based at least in part on at least one of an object type, a color, a size, a shape, a style, a brand, or a texture of the first object and the second visual similarity is based at least in part on at least one of an object type, a color, a size, a shape, a style, a brand, or a texture of the second object; and

present at least a portion of the second image, wherein the at least a portion of the second image-includes the third object.

2. The computing system of claim 1 , wherein the at least a portion of the second image is presented concurrently with the first image.

3. The computing system of claim 1 , wherein the instructions when executed further cause the computing system to at least:

for each of the plurality of objects:

index the object using a text-based retrieval technique such that the object is included in at least one index maintained in the data store.

4. The computing system of claim 1 , wherein the instructions when executed further cause the computing system to at least:

receive a third selection of a fourth object represented in the first image of the plurality of objects; and

wherein the second image further includes a fifth object having a third visual similarity to the fourth object, the third visual similarity being based at least in part on at least one of an object type, a color, a size, a shape, a style, a brand, or a texture of the fourth object.

5. The computing system of claim 1 , wherein the instructions when executed further cause the computing system to at least:

present on the first image, a graphical representation of a selection control identifying that an object included in the first image may be selected.

6. The computing system of claim 1 , wherein the instructions when executed further cause the computing system to at least:

receive a third selection of a fourth object of the plurality of objects from the first image;

determine that the second image includes a fifth object having a third visual similarity to the fourth object, the third visual similarity being based at least in part on at least one of an object type, a color, a size, a shape, a style, a brand, or a texture of the fourth object; and

wherein the presented at least a portion of the second image includes at least a portion of the fifth object.

7. The computing system of claim 1 , wherein the instructions, when executed by the at least one processor, further cause the computing system to at least:

determine a first similarity score corresponding to the first visual similarity of the first object and the third object;

receive a third selection of a fourth object of the plurality of objects from the first image;

determine a second similarity score corresponding to a third visual similarity of the fourth object and a fifth object included in the second image; and

determine a combined similarity score corresponding to the second image, the combined similarity score being based at least in part on a combination of the first similarity score and the second similarity score,

wherein the third visual similarity is based at least in part on at least one of an object type, a color, a size, a shape, a style, a brand, or a texture of the fourth object.

8. A computer-implemented method, comprising:

processing a first image that includes a plurality of objects to detect each of the plurality of objects;

for each of the plurality of objects, associating a token with the object;

receiving a first selection of a first location within the first image;

receiving a second selection of a second location within the first image;

determining a first token corresponding to a first object represented at the first location;

determining a second token corresponding to a second object represented at the second location;

querying at least one index maintained in a data store to determine, based at least in part on the first token and the second token, a plurality of additional images indicated in the at least one index, each of the plurality of additional images including a representation of an additional object having a first visual similarity to the first object and does not include an object having a second visual similarity to the second object, wherein the first visual similarity is based at least in part on at least one of an object type, a color, a size, a shape, a style, a brand, or a texture of the first object and the second visual similarity is based at least in part on at least one of an object type, a color, a size, a shape, a style, a brand, or a texture of the second object; and

presenting, concurrently with the first image, at least a portion of each of the plurality of additional images.

9. The computer-implemented method of claim 8 , wherein the first selection is at least one of a touch-based input on a display, a determined position of a user's gaze, or an input from an input component.

10. The computer-implemented method of claim 8 , further comprising:

determining that the first selection is a positive selection; and

determining that the second selection is a negative selection.

11. The computer-implemented method of claim 8 , further comprising:

presenting at least one selection control at a location on the first image to identify an object represented in the first image that is selectable.

12. The computer-implemented method of claim 8 , further comprising:

presenting adjacent each of the plurality of objects a selection control indicating that the respective object may be selected.

13. The computer-implemented method of claim 8 , further comprising:

determining for each of the plurality of additional images a respective similarity score;

ranking each of the plurality of additional images based at least in part on the respective similarity scores; and

wherein presenting includes presenting, concurrently with the first image, at least a portion of the plurality of additional images in an ordered arrangement based at least in part on the respective similarity scores.

14. The computer-implemented method of claim 13 , wherein each of the plurality of additional images includes a plurality of additional objects and the respective similarity score for each of the plurality of additional images includes a combination of a plurality of additional object similarity scores, each of the plurality of additional object similarity scores corresponding to a respective additional object similarity score for each of the plurality of additional objects.

15. The computer-implemented method of claim 8 , further comprising:

receiving a third selection of a second image from the plurality of additional images;

removing the presentation of the first image; and

presenting the second image.

16. A non-transitory computer-readable storage medium storing instructions that, when executed by at least one processor of a computing system, cause the computing system to at least:

receive a selection of a first representation of a first object included in a first image presented in a first window;

receive a second selection of a third representation of a third object included in the first image;

in response to the selection and the second selection, process a first region of the first image corresponding to the first object to form a first token representative of the first object and process a second region of the first image to form a second token representative of the third object;

query at least one index maintained in a data store to determine, based at least in part on the first token and the second token, a second image indicated in the at least one index, each of the second image including a second representation of a second object having a first visual similarity to the first object and does not include a representation of an object having a second visual similarity to the third object, wherein the first visual similarity is based at least in part on at least one of an object type, a color, a size, a shape, a style, a brand, or a texture of the first object and the second visual similarity is based at least in part on at least one of an object type, a color, a size, a shape, a style, a brand, or a texture of the second object; and

present, in a second window, at least a portion of the second image.

17. The non-transitory computer-readable storage medium of claim 16 , further comprising:

receive a third selection of a third representation of a third object included in the first image;

in response to the third selection, process a second region of the first image to form a second token; and

wherein the query of the at least one index is further to determine, based at least in part on the second token, the second representation such that the second representation includes a representation of an object having a third visual similarity to the third object, the third visual similarity being based at least in part on at least one of an object type, a color, a size, a shape, a style, a brand, or a texture of the third object.

18. The non-transitory computer-readable storage medium of claim 16 , further comprising:

indexing the first object in the at least one data store.

19. The non-transitory computer-readable storage medium of claim 16 , wherein the instructions when executed further cause the computing system to at least:

receive a third selection of the presented at least a portion of the second image; and

present the second image in the first window.

20. The non-transitory computer-readable storage medium of claim 16 , wherein the instructions when executed further cause the computing system to at least:

determine a third image that includes a third representation of a third object having a third visual similarity to the first object, the third visual similarity being based at least in part on at least one of an object type, a color, a size, a shape, a style, a brand, or a texture of the first object;

rank the second image and the third image based at least in part on a similarity score associated with each of the second image and the third image; and

present, in the second window, a ranked representation of at least a portion of the second image and the third image.