Image sharing with facial recognition models
View Patent ↗An image sharing server provides several ways of sharing images between users. After a user contributes images to the image sharing server, the user can interact with the image sharing server to identify and tag people in the images, share the images with other users, and organize the images into memory boxes. Memory boxes can also be shared between users, and multiple users can be granted the ability to add images to a shared memory box. In addition, the image sharing server can prompt a user to share his or her images with other users who contributed related images. The image sharing server also performs facial recognition to automatically identify people in images, and facial recognition models can be shared between users.
1. A computer-implemented method for managing and improving face recognition across user accounts in a computer network system, comprising:
storing a first account comprising a plurality of images for a first user in a database;
constructing, by one or more image servers, a first face model for a person using a first set of images in the first account, the first set of images each depicting the person's face;
storing a second account comprising a plurality of images for a second user in the database;
constructing, by the one or more image servers, a second face model for the same person using a second set of images in the second account, the second set of images each depicting the person's face, wherein number of images in the first set of images is different from number of images in the second set of images;
sharing the first face model with the second user account; and
recognizing, by the one or more image servers, faces of the person in the images in the second account at least in part based on the first face model.
2. The computer-implemented method of claim 1 , further comprising:
revising the second face model using the first face model to obtain a refined second face model; and
recognizing, by the one or more image servers, faces of the person in the images in the second account based on the refined second face model.
3. The computer-implemented method of claim 2 , wherein the second face model is revised while the first set of images in the first account are not shared with the second account.
4. The computer-implemented method of claim 1 , further comprising:
storing a tag of recognized people in association with each of the images in the first set of images and the second set of images, the tag indicating that the person is identified;
storing a first link to the first face model in association with each tag associated with each of the images in the first set of images; and
storing a second link to the second face model in association with each tag associated with each of the images in the second set of images.
5. The computer-implemented method of claim 4 , further comprising:
after the first face model is shared with the second user account, storing a third link to the first face model in association with each tag associated with each of the images in the second set of images.
6. The computer-implemented method of claim 1 , further comprising:
receiving a permission from the first user to share the first face model with the second user account.
7. A computer-implemented method for managing and improving face recognition across user accounts in a computer network system, comprising:
storing a first account comprising a plurality of images for a first user in a database;
constructing, by one or more image servers, a first face model for a person using a first set of images in the first account, the first set of images each depicting the person's face;
storing a second account comprising a plurality of images for a second user in the database;
constructing, by the one or more image servers, a second face model for the same person using a second set of images in the second account, the second set of images each depicting the person's face, wherein number of images in the first set of images is different from number of images in the second set of images;
sharing the first set of images with the second user account; and
revising the second face model using the first set of images to obtain a refined second face model.
8. The computer-implemented method of claim 7 , further comprising:
recognizing, by the one or more image servers, faces of the person in the images in the second account based on the refined second face model.
9. The computer-implemented method of claim 7 , further comprising:
storing a tag of recognized people in association with each of the images in the first set of images and the second set of images, the tag indicating that the person is identified;
storing a first link to the first face model in association with each tag associated with each of the images in the first set of images; and
storing a second link to the second face model in association with each tag associated with each of the images in the second set of images.
10. The computer-implemented method of claim 7 , further comprising:
receiving a permission from the first user to share the first set of images with the second user account.
11. A computer-implemented method for managing and improving face recognition across user accounts in a computer network system, comprising:
storing a first account comprising a plurality of images for a first user in a database;
constructing, by one or more image servers, a first face model for a person using a first set of images in the first account, the first set of images each depicting the person's face;
storing a second account comprising a plurality of images for a second user in the database, wherein a second set of images in the second account each depicts the person's face, wherein no face model has been developed for the person in the second account;
sharing the first face model with the second user account; and
recognizing, by the one or more image servers, faces of the person in the images in the second account at least in part based on the first face model.
12. The computer-implemented method of claim 11 , wherein the first face model is revised while the first set of images in the first account are not shared with the second account.
13. The computer-implemented method of claim 11 , further comprising:
storing a tag of recognized people in association with each of the images in the first set of images and the second set of images, the tag indicating that the person is identified;
storing a first link to the first face model in association with each tag associated with each of the images in the first set of images; and
after the first face model is shared with the second user account, storing a second link to the first face model in association with each tag associated with each of the images in the second set of images.
14. The computer-implemented method of claim 11 , further comprising:
receiving a permission from the first user to share the first face model with the second user account.