Incremental learning and validation of sequential predictors in video browsing application

Result code in IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21230%2F10%3A00175502" target="_blank" >RIV/68407700:21230/10:00175502 - isvavai.cz</a>
Result on the web
—
DOI - Digital Object Identifier
—

Result language
angličtina
Original language name
Incremental learning and validation of sequential predictors in video browsing application
Original language description
Loss-of-track detection (tracking validation) and automatic tracker adaptation to new object appearances are attractive topics in computer vision. We apply very efficient learnable sequential predictors in order to address both issues. Validation is doneby clustering of the sequential predictor responses. No aditional object model for validation is needed. The paper also proposes an incremental learning procedure that accommodates changing object appearance, which mainly improves the recall of the tracker/detector. Exemplars for the incremental learning are collected automatically, no user interaction is required. The aditional training examples are selected automatically using the tracker stability computed for each potential aditional training example. Coupled with a sparsely applied SIFT or SURF based detector the method is employed for object localization in videos. Our Matlab implementation scans videosequences up to eight times faster than the actual frame rate. A standard-lengt
Czech name
—
Czech description
—

Project
<a href="/en/project/GAP103%2F10%2F1585" target="_blank" >GAP103/10/1585: Advanced predictors for object detection and tracking in video</a><br>
Continuities
P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Publication year
2010
Confidentiality
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Article name in the collection
VISIGRAPP 2010: International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications
ISBN
978-989-674-028-3
ISSN
—
e-ISSN
—
Number of pages
8
Pages from-to
—
Publisher name
Institute for Systems and Technologies of Information, Control and Communication
Place of publication
Setúbal
Event location
Angers
Event date
May 17, 2010
Type of event by nationality
WRD - Celosvětová akce
UT code for WoS article
—

Similar results(10)