Show simple item record

dc.contributor.authorAHMADI, Aen_US
dc.contributor.authorPatras, Ien_US
dc.contributor.authorImage Processing (ICIP), 2016 IEEE International Conference onen_US
dc.date.accessioned2016-06-21T12:28:47Z
dc.date.available2016-05-06en_US
dc.date.issued2016-09-25en_US
dc.date.submitted2016-06-02T15:17:52.581Z
dc.identifier.urihttp://qmro.qmul.ac.uk/xmlui/handle/123456789/13000
dc.descriptionWe gratefully acknowledge the support of NVIDIA Corporation with the donation of the Tesla K40 GPU used for this research.
dc.descriptionWe gratefully acknowledge the support of NVIDIA Corporation with the donation of the Tesla K40 GPU used for this research.en_US
dc.description.abstractTraditional methods for motion estimation estimate the motion field F between a pair of images as the one that minimizes a predesigned cost function. In this paper, we propose a direct method and train a Convolutional Neural Network (CNN) that when, at test time, is given a pair of images as input it produces a dense motion field F at its output layer. In the absence of large datasets with ground truth motion that would allow classical supervised training, we propose to train the network in an unsupervised manner. The proposed cost function that is optimized during training, is based on the classical optical flow constraint. The latter is differentiable with respect to the motion field and, therefore, allows backpropagation of the error to previous layers of the network. Our method is tested on both synthetic and real image sequences and performs similarly to the state-of-the-art methods.en_US
dc.publisherIEEE Xploreen_US
dc.rightshttp://arxiv.org/abs/1601.06087
dc.subjectMotion Estimationen_US
dc.subjectConvolutional Neural Networken_US
dc.subjectUnsupervised Trainingen_US
dc.titleUNSUPERVISED CONVOLUTIONAL NEURAL NETWORKS FOR MOTION ESTIMATIONen_US
dc.typeConference Proceeding
dc.identifier.doi10.1109/ICIP.2016.7532634en_US
pubs.notesNot knownen_US
pubs.publication-statusPublished onlineen_US
pubs.publisher-urlhttp://ieeexplore.ieee.org/xpl/mostRecentIssue.jsp?punumber=7527113en_US
dcterms.dateAccepted2016-05-06en_US


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record