Show simple item record

dc.contributor.advisorBoots, Byron
dc.contributor.authorShaban, Amirreza
dc.date.accessioned2020-09-08T12:44:48Z
dc.date.available2020-09-08T12:44:48Z
dc.date.created2020-08
dc.date.issued2020-05-17
dc.date.submittedAugust 2020
dc.identifier.urihttp://hdl.handle.net/1853/63599
dc.description.abstractDeep Neural Networks are powerful at solving classification problems in computer vision. However, learning classifiers with these models requires a large amount of labeled training data, and recent approaches have struggled to adapt to new classes in a data-efficient manner. On the other hand, the human brain is capable of utilizing already known knowledge in order to learn new concepts with fewer examples and less supervision. Many meta-learning algorithms have been proposed to fill this gap but they come with their practical and theoretical limitations. We review the well-known bi-level optimization as a general framework for few-shot learning and hyperparameter optimization and discuss the practical limitations of computing the full gradient. We provide theoretical guarantees for the convergence of the bi-level optimization using the approximated gradients computed by the truncated back-propagation. In the next step, we propose an empirical method for few-shot semantic segmentation: instead of solving the inner optimization, we propose to directly estimate its result by a general function approximator. Finally, we will discuss extensions of this work with the focus on weakly-supervised object detection when full supervision is not available for the few training examples.
dc.format.mimetypeapplication/pdf
dc.language.isoen_US
dc.publisherGeorgia Institute of Technology
dc.subjectFew-shot learning
dc.subjectLow-shot learning
dc.subjectBi-level optimization
dc.subjectFew-shot semantic segmentation
dc.subjectVideo object segmentation
dc.subjectWeakly-supervised few-shot object detection
dc.titleLow-shot learning for object recognition, detection, and segmentation
dc.typeDissertation
dc.description.degreePh.D.
dc.contributor.departmentInteractive Computing
thesis.degree.levelDoctoral
dc.contributor.committeeMemberHays, James
dc.contributor.committeeMemberBatra, Dhruv
dc.contributor.committeeMemberKira, Zsolt
dc.contributor.committeeMemberLi, Fuxin
dc.date.updated2020-09-08T12:44:48Z


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record