Probabilistic Data Association Methods for Tracking Multiple and Compound Visual Objects

DMCA

Probabilistic Data Association Methods for Tracking Multiple and Compound Visual Objects (2000)

Cached

Download Links

Other Repositories/Bibliography

DBLP

by Christopher Rasmussen , Gregory D. Hager

Venue:	IEEE Transactions on Pattern Analysis and Machine Intelligence
Citations:	154 - 2 self

BibTeX

@ARTICLE{Rasmussen00probabilisticdata,
    author = {Christopher Rasmussen and Gregory D. Hager},
    title = {Probabilistic Data Association Methods for Tracking Multiple and Compound Visual Objects},
    journal = {IEEE Transactions on Pattern Analysis and Machine Intelligence},
    year = {2000},
    volume = {23},
    pages = {560--576}
}

OpenURL

Abstract

We describe a framework that explicitly reasons about data association to improve tracking performance in many difficult visual environments. A hierarchy of tracking strategies results from ascribing ambiguous or missing data to: (1) noise-like visual occurrences; (2) persistent, known scene elements (i.e. other tracked objects); or (3) persistent, unknown scene elements. First, we introduce a randomized tracking algorithm adapted from an existing probabilistic data association filter (PDAF) that is resistant to clutter and follows agile motion. The algorithm is applied to three different tracking modalities -- homogeneous regions, textured regions, and snakes -- and extensibly defined for straightforward inclusion of other methods. Second, we add the capacity to track multiple objects by adapting to vision a joint PDAF which oversees correspondence choices between same-modality trackers and image features. We then derive a related technique that allows mixed tracker modalities and handles object...

Keyphrases

probabilistic data association method compound visual object modality homogeneous region image feature multiple object joint pdaf randomized tracking algorithm probabilistic data association filter correspondence choice strategy result agile motion scene element mixed tracker modality straightforward inclusion unknown scene element noise-like visual occurrence related technique data association many difficult visual environment tracked object same-modality tracker

DMCA