System and method for predictive curation, production infrastructure, and personal content assistant
Data points, calendar entries, trends, and behavioral patterns may be used to predict and pre-emptively build digital and printable products with selected collections of images without the user's active participation. The collections are selected from files on the user's device, cloud-based photo library, or other libraries shared among other individuals and grouped into thematic products. Based on analysis of the user's collections and on-line behaviors, the system may estimate types and volumes of potential media-centric products, and the resources needed for producing and distributing such media-centric products for a projected period of time. A user interface may take the form of a “virtual curator”, which is a graphical or animated persona for augmenting and managing interactions between the user and the system managing the user's stored media assets. The virtual curator can assume one of many personas with each user and can interact with the user via text/audio messaging.
1. A computer-implemented method configured to be performed on a networked computational device comprising a processor, the method comprising:
using the processor to verify an identity of a user account;
using the processor to identify a workflow trigger, wherein the workflow trigger comprises receiving multimedia assets at the networked computational device or a remote database;
using the processor to extract metadata associated with the multimedia assets;
using the processor to assign semantic tags to each multimedia asset based on the extracted metadata;
using the processor to sort the multimedia assets into groups, wherein each group corresponds to a semantic event and each semantic event is derived from the semantic tags;
using the processor to sort the multimedia assets in each group into thematic subgroups;
using the processor to assign group tags and thematic tags to the multimedia assets, wherein the group tags and thematic tags correspond to the groups and the thematic subgroups into which the multimedia assets are sorted;
weighing the semantic tags, the group tags, and the thematic tags based on a frequency of occurrence of each tag in association with the multimedia assets;
ranking the multimedia assets based on a weight of the semantic tags, the group tags, and the thematic tags associated with each multimedia asset;
identifying an important image, wherein the important image is the highest ranked multimedia asset; and
preparing a virtual version of a customized media-centric product to be offered to the user, wherein the customized media-centric product incorporates the important image.
2. The method of claim 1 , further comprising:
extracting user-preferred thematic tags from the user account, wherein weighing the semantic tags, the group tags, and the thematic tags comprises weighing the user-preferred thematic tags most heavily.
3. The method of claim 1 , further comprising:
using the processor to assess accuracy of the semantic tags, the group tags, and the thematic tags by comparing the semantic tags, the group tags, and the thematic tags to ground truth data.
4. The method of claim 3 , further comprising:
using the processor to run a deep learning model, wherein the deep learning model uses the assessment of accuracy of the semantic tags, the group tags, and the thematic tags to improve tagging models for assigning the semantic tags, the group tags, and the thematic tags.
5. The method of claim 1 , further comprising:
using the processor to analyze the semantic tags, the group tags, and the thematic tags to determine event boundaries within the multimedia assets.
6. The method of claim 1 , wherein using the processor to sort the multimedia assets into groups comprises forming groups based on location metadata in combination with ontological reasoning.