Automated semantic inference of visual features and scenes
US-10719744-B2 · Jul 21, 2020 · US
US12093303B2 · US · B2
| Field | Value |
|---|---|
| Publication number | US-12093303-B2 |
| Application number | US-202318139772-A |
| Country | US |
| Kind code | B2 |
| Filing date | Apr 26, 2023 |
| Priority date | Oct 10, 2013 |
| Publication date | Sep 17, 2024 |
| Grant date | Sep 17, 2024 |
A practical reading order for non-experts. Skip the full description unless you need deep technical detail.
What the patent document calls the invention.
A short plain-language summary of the technical disclosure.
Who owns or filed the patent and who is credited as inventor.
Filing, priority, publication, and grant dates set the timeline.
The legal scope of protection — read this for what is actually claimed.
Technology tags used to group this patent with similar filings.
Prior art links and similar publications in this corpus.
Official abstract text for this publication.
A method and apparatus for displaying digital photos on a digital display with a network connection module. The method includes identifying two different photos in a digital photo collection that include a matching photo content, and displaying the two different photos simultaneously in a split screen on the digital display. The two different photos can be matched from related photo clusters. An integrated camera can be used to automatically determine an identity of a frame viewer to determine photos to display, and can capture gesture-based feedback. The photos can be filtered and cropped at the receiver side.
Opening claim text (preview).
What is claimed is: 1. A method of displaying digital photos on a digital picture frame with a digital display screen and a network connection module, the method comprising: connecting to a first interaction application of a first electronic device of a first user through the network connection module, the first electronic device including a first digital photo collection; connecting to a second interaction application of a second electronic device of a second user through the network connection module, the second electronic device including a second digital photo collection; automatically identifying and selecting two different photos that include a matching photo content from the first and/or the second selected digital photo clusters; and displaying the two different photos simultaneously on the digital display screen, each in one of two equal screen sections of the digital display screen, wherein a first of the two different photos is in a first of the two equal screen sections, and a second of the two different photos is in an adjacent second of the two equal screen sections; wherein the matching photo content is a same person or persons, the two different photos are photographically captured in two different calendar years, the two different calendar years are separated by more than one additional calendar year, and further comprising: automatically clustering from the first digital photo collection, via automated facial recognition, photos over a multi-year timeframe of the person to obtain clustered photos of the person, wherein the clustering comprises: preparing a plurality of sub-clusters each including digital photos of the person at a similar time and/or age; extracting photo facial features within each of the plurality of sub-clusters that comprise a predetermined limited change over the timeframe to form a cluster representation of the person for each of the plurality of sub-clusters; comparing cluster representations between the plurality of sub-clusters to determine related sub-clusters of the person over the timeframe as a function of the cluster representations; linking or grouping related sub-clusters to form the clustered photos of the person for display; and automatically selecting a first of the two different photos from a first of the sub-clusters and a second of the two different photos from the second of the sub-clusters. 2. The method of claim 1 , wherein the first of the two different photos is a baby photo of a person in a first calendar year and the second of the two different photos is of the person at another age separated by the more than one additional calendar year; and further comprising: changing the second of the of the two different photos with a slideshow of the person at other ages, wherein the baby photo is not changed on the first of the two equal sections. 3. The method of claim 1 , wherein each of the two different photos further includes one of two different second persons, wherein the different second persons are related family members. 4. The method of claim 3 , wherein the two different photos are automatically downloaded from one or more digital photo collections of the two different second persons. 5. The method of claim 1 , further comprising: filtering each of the sub-clusters as a function of quality, wherein photos of a predetermined low quality are removed from the sub-clusters; and filtering the each of the sub-clusters as a function of a further content, wherein photos including predetermined people, activities and/or locations are removed from the sub-clusters. 6. The method of claim 1 , wherein the two different photos are displayed in a slideshow of digital photos on the digital display screen, wherein a plurality of displayed slideshow photos is automatically downloaded from electronic devices and/or social media accounts of a plurality of community members. 7. The method of claim 1 , wherein the digital picture frame automatically determines an identity of a viewer of the digital picture frame as the first user. 8. The method of claim 1 , further comprising: automatically identifying preferred photos or restricted photos in the sub-clusters as a function of user information. 9. The method of claim 7 , further comprising: detecting an electronic device of the viewer in proximity to the digital picture frame; and automatically scanning and clustering digital photos of the electronic device.
Detection; Localisation; Normalisation · CPC title
in albums, collections or shared content, e.g. social network photos or video · CPC title
based on graphs, e.g. graph cuts or spectral clustering · CPC title
using clustering, e.g. of similar faces in social networks · CPC title
by ranking or filtering the set of features, e.g. using a measure of variance or of feature cross-correlation · CPC title
Related publications grouped by family.
Answers are generated from the same data shown on this page.