We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

eess.SY

Change to browse by:

References & Citations

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Electrical Engineering and Systems Science > Systems and Control

Title: Incorporating Data Uncertainty in Object Tracking Algorithms

Abstract: Methodologies for incorporating the uncertainties characteristic of data-driven object detectors into object tracking algorithms are explored. Object tracking methods rely on measurement error models, typically in the form of measurement noise, false positive rates, and missed detection rates. Each of these quantities, in general, can be dependent on object or measurement location. However, for detections generated from neural-network processed camera inputs, these measurement error statistics are not sufficient to represent the primary source of errors, namely a dissimilarity between run-time sensor input and the training data upon which the detector was trained. To this end, we investigate incorporating data uncertainty into object tracking methods such as to improve the ability to track objects, and particularly those which out-of-distribution w.r.t. training data. The proposed methodologies are validated on an object tracking benchmark as well on experiments with a real autonomous aircraft.
Comments: For associated video, see this https URL
Subjects: Systems and Control (eess.SY); Computer Vision and Pattern Recognition (cs.CV)
Cite as: arXiv:2109.10521 [eess.SY]
  (or arXiv:2109.10521v1 [eess.SY] for this version)

Submission history

From: Anish Muthali [view email]
[v1] Wed, 22 Sep 2021 05:30:46 GMT (1299kb,D)

Link back to: arXiv, form interface, contact.