Scalable full flow with learned binary descriptors
The result's identifiers
Result code in IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21230%2F17%3A00315701" target="_blank" >RIV/68407700:21230/17:00315701 - isvavai.cz</a>
Result on the web
<a href="http://dx.doi.org/10.1007/978-3-319-66709-6_26" target="_blank" >http://dx.doi.org/10.1007/978-3-319-66709-6_26</a>
DOI - Digital Object Identifier
<a href="http://dx.doi.org/10.1007/978-3-319-66709-6_26" target="_blank" >10.1007/978-3-319-66709-6_26</a>
Alternative languages
Result language
angličtina
Original language name
Scalable full flow with learned binary descriptors
Original language description
We propose a method for large displacement optical flow in which local matching costs are learned by a convolutional neural network (CNN) and a smoothness prior is imposed by a conditional random field (CRF). We tackle the computation- and memory-intensive operations on the 4D cost volume by a min-projection which reduces memory complexity from quadratic to linear and binary descriptors for efficient matching. This enables evaluation of the cost on the fly and allows to perform learning and CRF inference on high resolution images without ever storing the 4D cost volume. To address the problem of learning binary descriptors we propose a new hybrid learning scheme. In contrast to current state of the art approaches for learning binary CNNs we can compute the exact non-zero gradient within our model. We compare several methods for training binary descriptors and show results on public available benchmarks.
Czech name
—
Czech description
—
Classification
Type
D - Article in proceedings
CEP classification
—
OECD FORD branch
10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)
Result continuities
Project
—
Continuities
N - Vyzkumna aktivita podporovana z neverejnych zdroju
Others
Publication year
2017
Confidentiality
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů
Data specific for result type
Article name in the collection
39th German Conference on Pattern Recognition
ISBN
978-3-319-66708-9
ISSN
0302-9743
e-ISSN
—
Number of pages
12
Pages from-to
321-332
Publisher name
Springer, Cham
Place of publication
—
Event location
Basel
Event date
Sep 12, 2017
Type of event by nationality
WRD - Celosvětová akce
UT code for WoS article
—