Learning Multi-View Correspondences via Subspace-Based Temporal Coincidences
2013 (English)In: Proceeding Scandinavian Conference on Image Analysis, 2013, Springer Berlin/Heidelberg, 2013, 456-467 p.Conference paper (Other academic)
In this work we present an approach to automatically learn pixel correspondences between pairs of cameras. We build on the method of Temporal Coincidence Analysis (TCA) and extend it from the pure temporal (i.e. single-pixel) to the spatiotemporal domain. Our approach is based on learning a statistical model for local spatiotemporal image patches, determining rare, and expressive events from this model, and matching these events across multiple views. Accumulating multi-image coincidences of such events over time allows to learn the desired geometric and photometric relations. The presented method also works for strongly different viewpoints and camera settings, including substantial rotation, and translation. The only assumption that is made is that the relative orientation of pairs of cameras may be arbitrary, but fixed, and that the observed scene shows visual activity. We show that the proposed method outperforms the single pixel approach to TCA both in terms of learning speed and accuracy.
Place, publisher, year, edition, pages
Springer Berlin/Heidelberg, 2013. 456-467 p.
Lecture Notes in Computer Science, ISSN 0302-9743 (print), 1611-3349 (online) ; 7944
Engineering and Technology
IdentifiersURN: urn:nbn:se:liu:diva-90921DOI: 10.1007/978-3-642-38886-6_43ISBN: 978-3-642-38885-9 (print)ISBN: 978-3-642-38886-6 (online)OAI: oai:DiVA.org:liu-90921DiVA: diva2:615201
18th Scandinavian Conference on Image Analysis (SCIA 2013), 17-20 June 2013, Espoo, Finland
FundereLLIIT - The Linköping‐Lund Initiative on IT and Mobile Communications