System and method for use of images with recognition analysis
View Patent ↗An index is provided that holds information about each image content item in a collection of items, For each image content item, a first information item identifying the image content item and its location on a network, and at least one of (i) a second information item identifying a signature value of an object in the image content, or (ii) identification of a recognized object in the image content.
1. A method performed by a system of one or more computers for selecting image content, the method comprising:
performing image analysis on a collection of image content items to obtain information about each image content item, including one or more of(i) determination of one or more objects in individual image content item, (ii) determination of a signature value for each determined object;
detecting content that is to be used for image selection;
providing, to a user, one or more image selection results from the collection of image content items using the detected content and the information obtained from the image analysis;
receiving, from the user, feedback on a first image selection result, the feedback referring to first image content of the first image selection result; and
providing, to the user, one or more additional image selection results using the detected content, the information obtained from the image analysis, and the feedback on the first image selection result;
wherein receiving feedback on a first image selection result comprises: causing a slider user interface element to be displayed with the first image content; and receiving, as a result of the user interacting with the slider, a specified change to a shape, color, pattern, or style within the first image content.
2. The method of claim 1 , wherein:
receiving feedback on a first image selection result comprises causing a color picker user interface element to be displayed with the first image content and receiving, as a result of the user interacting with the color picker, a particular color; and
providing the one or more additional image selection results comprises performing a search operation to identify one or more additional image content items containing color that matches the particular color.
3. The method of claim 1 , wherein:
receiving feedback on a first image selection result comprises receiving, from the user, an identification of a key region within the first image content and determining one or more local features within the key region; and
providing the one or more additional image selection results comprises performing a search operation using the local features.
4. The method of claim 1 , wherein detecting content for image selection comprises:
receiving a user image content item from the user for selection;
identifying an object of the user image content item;
forming a criteria for providing the image selection results based on the identified object.
5. The method of claim 4 , wherein forming the criteria includes forming a text string criteria that is based on identification of the object.
6. The method of claim 1 , wherein detecting content for image selection comprises:
receiving a user text content for selecting image content items from the user; and
forming a criteria for providing the image selection results based on the text content.
7. The method of claim 1 , wherein detecting content for image selection comprises:
receiving a user image content item from the user for selection;
identifying text data associated with the image content item; and
forming a criteria for providing the image selection results based on the text data associated with the image content item.
8. The method of claim 1 , wherein detecting content for image selection comprises receiving an identifier of image content of a particular merchandise object, and wherein providing the image selection results comprises providing image selection results of different image content items of different merchandise objects based on a determination that the different image content items are similar to the particular merchandise object.
9. The method of claim 1 , comprising providing a link with each of the different image content items of different merchandise objects, wherein the link is selectable to enable the user to purchase the merchandise objects from a network site operated by a merchant.
10. A system of one or more computers configured to perform operations comprising:
performing image analysis on a collection of image content items to obtain information about each image content item, including one or more of (i) determination of one or more objects in individual image content item, (ii) determination of a signature value for each determined object;
detecting content that is to be used for image selection;
providing, to a user, one or more image selection results from the collection of image content items using the detected content and the information obtained from the image analysis;
receiving, from the user, feedback on a first image selection result, the feedback referring to first image content of the first image selection result; and
providing, to the user, one or more additional image selection results using the detected content, the information obtained from the image analysis, and the feedback on the first image selection result; wherein receiving feedback on a first image selection result comprises:
causing a slider user interface element to be displayed with the first image content; and
receiving, as a result of the user interacting with the slider, a specified change to a shape, color, pattern, or style within the first image content.
11. The system of claim 10 , wherein:
receiving feedback on a first image selection result comprises causing a color picker user interface element to be displayed with the first image content and receiving, as a result of the user interacting with the color picker, a particular color; and
providing the one or more additional image selection results comprises performing a search operation to identify one or more additional image content items containing color that matches the particular color.
12. The system of claim 10 , wherein:
receiving feedback on a first image selection result comprises receiving, from the user, an identification of a key region within the first image content and determining one or more local features within the key region; and
providing the one or more additional image selection results comprises performing a search operation using the local features.
13. The system of claim 10 , wherein detecting content for image selection comprises:
receiving a user image content item from the user for selection;
identifying an object of the user image content item;
forming a criteria for providing the image selection results based on the identified object.
14. The system of claim 13 , wherein forming the criteria includes forming a text string criteria that is based on identification of the object.
15. The system of claim 10 , wherein detecting content for image selection comprises:
receiving a user text content for selecting image content items from the user; and
forming a criteria for providing the image selection results based on the text content.
16. The system of claim 10 , wherein detecting content for image selection comprises:
receiving a user image content item from the user for selection;
identifying text data associated with the image content item; and
forming a criteria for providing the image selection results based on the text data associated with the image content item.
17. The system of claim 10 , wherein detecting content for image selection comprises receiving an identifier of image content of a particular merchandise object, and wherein providing the image selection results comprises providing image selection results of different image content items of different merchandise objects based on a determination that the different image content items are similar to the particular merchandise object.
18. The system of claim 10 , the operations comprising providing a link with each of the different image content items of different merchandise objects, wherein the link is selectable to enable the user to purchase the merchandise objects from a network site operated by a merchant.