Blog Archive: 2010

Active capture at ACM MM 2010


FXPAL has a few papers appearing at the upcoming ACM Multimedia Conference in Firenze, Italy.  Among them is NudgeCam, which was recently featured in an article on MIT’s Technology Review as noted previously on this very blog.

NudgeCam is an experiment in “active capture”. Media capture (in this case, photos and videos) is enhanced by providing a template of elements to capture and also real-time interactive tips to aid the quality of each shot or clip.  The template allows the author to insure that essential story components are captured, and the realtime feedback helps insure that the parts are of high quality. Together the creation of high quality result is streamlined.

The author, Scott Carter, will be presenting this work on Tuesday, October 26th in Session S1 at ACM Multimedia in Firenze, Italy.

See you there!

FXPAL is hiring

on Comments (2)

FXPAL has immediate openings for regular full time employees and visiting scientists with research interests in the following areas: cloud computing, computer vision, multimedia applications, location-aware and event-processing applications, interactive documents, mixed reality environments, and database systems.

Candidates should be interested in working on practical applications in a collaborative setting. Regular full time positions require a Ph.D. in Computer Science or related field and strong development skills. Visiting Scientist positions require an extensive record of research contributions and experience.

Update: See the FXPAL web site to apply

Nudging the world toward better pictures and video

on Comments (2)

An excellent article on FXPAL’s NudgeCam application recently appeared in MIT’s Technology Review. NudgeCam encapsulates standard video capture heuristics, such as how to frame a face and good brightness characteristics, in order to provide guidance to users as they are taking video, using image analysis techniques such as face recognition,  as to how to adjust the camera to improve the video capture.

For its size, FXPAL has surprising breadth and variety of expertise. The NudgeCam work resulted from a collaboration between Scott Carter, whose expertise is in mobile and ubiquitous computing,  and John Doherty, our multimedia specialist, who knows all the standard video capture heuristics and many more. John Adcock brought image analysis techniques to the team, and 2009 FXPAL summer intern Stacy Branham contributed her human-computer interaction expertise.

A different application, also developed at FXPAL, supports rephotography in an industrial setting. Rephotography is the art of taking a photograph from the same location and angle as a previous photograph. Continue Reading

Reverted Indexing

on Comments (8)

Traditional interactive information retrieval systems function by creating inverted lists, or term indexes. For every term in the vocabulary, a list is created that contains the documents in which that term occurs and its relative frequency within each document. Retrieval algorithms then use these term frequencies alongside other collection statistics to identify the matching documents for a query.

In a paper to be published at CIKM 2010, Jeremy Pickens, Matt Cooper and  I describe a way of using the inverted index to associate document ids with the queries that retrieve them. Our approach combines the inverted index with the notion of retrievability to create an efficient query expansion algorithm that is useful for a number of applications, including relevance feedback. We call this kind of index a reverted index because rather than mapping terms onto documents, it maps document ids onto queries that retrieved the associated documents.

Continue Reading


on Comments (6)

While many of the systems we build at FXPAL are either deployed internally or transferred to our parent company, in some cases we get to deploy them in the real world. This week, we released TalkMiner, a system for indexing and searching video of lecture broadcasts. We’ve indexed broadcasts from a variety of sources, including the U.C. Berkeley webcast.berkeley site, the site, and various channels on YouTube, including Google Tech Talks, Stanford University, MIT Open Courseware, O’Reilly Media, TED Talks, and NPTEL Indian Institute of Technology.

But all of these videos are already indexed by web search engines, you say; why do we need TalkMiner?

Continue Reading

Session-based search slides

on Comments (3)

Here are the slides of the presentation I gave at the IIiX 2010 conference. I presented work done in collaboration with Jeremy Pickens on session-based search. The paper is here; the talk highlights some of the theoretical considerations and gives some examples of the new interface we’re building.

Continue Reading

Session-based search

on Comments (14)

Exploratory search often takes place over time. Searchers may run multiple queries to understand the collection, to refine their information needs, or to explore various aspects of the topic of interest. Many web search engines keep a history of a user’s actions: Bing makes that history readily available for backtracking, and all major search engines presumably use the click-through history of search results to affect subsequent searches. Yahoo Search Pad diagnoses exploratory search situations and switches to a more elaborate note-taking mode to help users manage the found information.

But none of these approaches makes it easy for a searcher to manage an on-going exploratory search. So what could be done differently? We explore this topic in a paper we’ll be presenting at the IIiX 2010 conference this August. Our paper reviews the literature on session-based search, and proposes a framework for designing interactions around information seeking. This framework uses the structure of the process of exploratory search to help searchers reflect on their actions and on the retrieved results. It treats queries, terms, metadata, documents, sets of queries, and sets of documents as first-class objects that the user can manipulate, and describes how information seeking context can be preserved across these transitions.

Continue Reading

FXPAL turns 15

on Comments (1)

Today is FXPAL’s 15th anniversary. While we’re young by the standards of research labs (IBM Watson was founded in 1945, (Xerox) PARC in 1970,  IBM Almaden in 1986, and MSR in 1991), we’ve managed to accumulate a pretty good record for our size. With an annual staff of 20-25 PhD-level researchers and a bunch of summer interns, we’ve consistently produced about 30 research publications a year on a variety of research topics ranging from multimedia to HCI to information retrieval. While no single post can do justice to the great work of so many people, here are some highlights.

Continue Reading

ReBoard presentation at CHI 2010

on Comments (1)

Stacy Branham who was an intern with us last summer gave an excellent talk at CHI 2010 about the study that she ran of how people use ReBoard. I’ve written about the study before, and the papers are available here and here. But the slides are interesting in their own right, and tell a complementary story.

First, there are the CHI Madness slides:

Let the slides auto-advance for the best effect.

Continue Reading