The Community/Individual Cycle: Extracting Knowledge, Creating Applications and Understanding Motivations in Public Photo Collections

Speaker: Mor Naaman , Yahoo! Research Berkeley

Date: Friday, February 23, 2007

Time: 1:30 PM to 2:30 PM

Refreshments: 1:15 PM

Public: Yes

Location: Star Seminar Room (D463)

Event Type:

Room Description:

Host: Rob Miller, MIT CSAIL

Contact: Michael Bernstein,

Relevant URL:

Speaker URL: None

Speaker Photo:

Reminders to:,

Reminder Subject: TALK: The Community/Individual Cycle

Public collections of shared photographs can expose large-scale photo-taking
patterns. These patterns provide far richer information than can be derived
from individual images or photo sets. I will first describe how we extract
meaningful information from public collections of "geotagged" images on
Flickr. This type of information allows us, for example, to produce
summaries and visualizations of the photo collections, making them more
accessible and easier to browse and understand, in a manner that improves
rather than degrades with the addition of more photos.

I will then describe how the same type of information can also be used to
improve any user's photo experience, using new applications such as ZoneTag.
ZoneTag is a mobile client that enables capture and upload of cameraphone
photographs. ZoneTag supports annotation of the captured media via
context-based 'tag suggestions', partially derived from patterns in Flickr's
public photo collection. ZoneTag thus forms the basis for a rich personal
and social media retrieval/organization and sharing system.

I will wrap up the talk by describing two user studies that investigated
user motivations in photo collections. We looked at the factors that come
into play when people making privacy and tagging decisions in Flickr and
ZoneTag, thus contributing data back to the community. We can then use this
data in extracting the meaningful patterns, as described in paragraph 1.

Mor Naaman is a research scientist at Yahoo! Research Berkeley. His research
focuses on context-based tools and algorithms for interacting with media.
Mor has a Ph.D. in Computer Science from Stanford University. His research
in the Stanford Infolab also focused on management of digital photographs,
thereby allowing (and requiring!) him to take photos throughout his working
life. In previous careers, Mor was a professional basketball player as well
as a software developer and a college radio DJ.

Research Areas:

Impact Areas:

See other events that are part of the HCI Seminar Series Spring 2007.

Created by Linda L. Julien Email at Wednesday, June 19, 2013 at 6:22 AM.