Reverse nearest neighbor queries are useful in identifying objects that are of significant influence or importance. Existing methods either rely on pre-computation of nearest neighbor distances, do not scale well with high dimensionality, or do not produce exact solutions. In this work we motivate and investigate the problem of reverse nearest neighbor… (More)
We introduce Regression Databases (REDB) to formalize and automate probabilistic querying using sparse learning sets. The REDB data model involves observation data, learning set data, views definitions, and a regression model instance. The observation data is a collection of relational tuples over a set of attributes; the learning data set involves a subset… (More)
Known Item Search (KIS) is a specialized task of the general multimedia search problem. It describes the scenario where a user has previously seen a video and wants to find it again in a large collection using a text description. While there exists only one correct answer to a query (or topic), the goal is to return a ranked list of videos most likely to… (More)
This paper describes KB Video Retrieval's participation in the TREC Video Retrieval Evaluation for 2010. This year we submitted results for the Semantic Indexing, Known-item Search, Instance Search, and Event Detection in Internet Multimedia tasks. Our goal this year was to evaluate ranking strategies and expand our knowledge based approach to a variety of… (More)
This paper describes the Knowledge Base multimedia retrieval system for the TRECVID 2008 evaluation. Our focus this year is on query analysis and the creation of a topic knowledge base using external knowledge base information.
—Multimedia retrieval suffers from the lack of common feature representation between a text based query and the visual content of a video repository. One approach to bridging this representation gap is known as query-by-concept, where a query and video are mapped into a common semantic feature space. One of the challenges with using semantic concepts for… (More)
Etter Solutions Research Group participated in the TRECVID conference for the first time in 2007. We submitted five runs in the area of fully automatic search. F_A_1_ESRG_A1_1 The first run is a required baseline, using only ASR/MT features. A sliding window of three shots is used to create a " bag of words " representation of the shot. F_A_1_ESRG_KN2_2 The… (More)