Browse Prior Art Database

Adaptive media file selection for social network sharing Disclosure Number: IPCOM000239780D
Publication Date: 2014-Dec-01
Document File: 2 page(s) / 27K

Publishing Venue

The Prior Art Database


Disclosed is a method that enables automatic selection of pictures being shared in a social network that are of interest to individual viewers

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 52% of the total text.

Page 01 of 2

Adaptive media file selection for social network sharing
People take pictures or videos during vacation to capture the experience and memory. Tens or hundreds of pictures/videos can be taken on a long trip. Users may have difficulty choosing which pictures or videos to share with their friends/family. If the user uploads all the pictures to be shared in a social network, viewers may not have patient to go through all of them or consider all of the pictures interesting to them.


We proposed a means to enhance the viewing experience from the viewers' perspective. We propose a dynamic subset of pictures to be automatically selected to show the viewers based on each viewer's interest. The user (content provider) can upload the entire set of pictures/videos to a social network. When his/her friend or family member views the photo album, the system automatically selects a subset of pictures that fits the viewer's interests. In addition, the number of pictures selected can be adjusted by the viewing habit of the viewer or a predictive time that the viewer will spent in the social network.

For each viewer, two levels of filters can be defined:
Level 1 filter is related to the viewer's immediate interests at a given point of time.

Level 2 filter is related to the activities of the circle of common friends between the viewer and the content provider at a given point of time.


Mary spent a week in Japan. She visited a lot of temples, a botanical garden, an aquarium, Universal studio, fish market, and she tried out a lot of food including kobe beef, seafood etc. She has taken 100 pictures for the trip.

Here are some of her friends on Facebook: Wendy - vegetarian, love animals
Jenny - a foodie, loves movies
Peter - spiritual
Kerrie - loves flowers

Step 1 1:

: Pictures categorization :

When a user takes a picture, his/her picture can be automatically tagged by the location it was taken. Categorization of pictures can be done by inference of location. For example, Mary visited a zoo. Her cell phone has GPS capability to tell her location is a zoo. The pictures taken can be tagged with the word "zoo" as tag. Her pictures taken at the zoo can be categorized as "animals" which are inferred by the location tag. The system can perform a search query on-line to discover the location type if its GPS location has not already associated with a location type.

Step 2 2:

: Pictures posting :

The user creates a photo album for a trip in a social network. The user posts all the pictures/videos of the trip in the new album. The user is free to post hundreds of pictures.

Step 3 3:

: Pictures are shared among the content provider '
'''s friends or family

s friends or family . Each viewer '
'''s profile is analyzed and summarized

s profile is analyzed and summarized .

Information including :
-The viewer's profile description (age, interests/hobbies, locations, age, sex, etc)

- The viewer's liking (identify types of content that att...