Low-shot Visual Recognition

dc.contributor.authorPemula, Lathaen
dc.contributor.committeechairBatra, Dhruven
dc.contributor.committeememberParikh, Devien
dc.contributor.committeememberAbbott, A. Lynnen
dc.contributor.departmentElectrical and Computer Engineeringen
dc.date.accessioned2016-10-25T08:00:37Zen
dc.date.available2016-10-25T08:00:37Zen
dc.date.issued2016-10-24en
dc.description.abstractMany real world datasets are characterized by having a long tailed distribution, with several samples for some classes and only a few samples for other classes. While many Deep Learning based solutions exist for object recognition when hundreds of samples are available, there are not many solutions for the case when there are only a few samples available per class. Recognition in the regime where the number of training samples available for each class are low, ranging from 1 to couple of tens of examples is called Lowshot Recognition. In this work, we attempt to solve this problem. Our framework is similar to [1]. We use a related dataset with sufficient number (a couple of hundred) of samples per class to learn representations using a Convolutional Neural Network (CNN). This CNN is used to extract features of the lowshot samples and learn a classifier . During representation learning, we enforce the learnt representations to obey certain property by using a custom loss function. We believe that when the lowshot sample obey this property the classification step becomes easier. We show that the proposed solution performs better than the softmax classifier by a good margin.en
dc.description.degreeMaster of Scienceen
dc.format.mediumETDen
dc.identifier.othervt_gsexam:9048en
dc.identifier.urihttp://hdl.handle.net/10919/73321en
dc.publisherVirginia Techen
dc.rightsIn Copyrighten
dc.rights.urihttp://rightsstatements.org/vocab/InC/1.0/en
dc.subjectVisual Recognitionen
dc.subjectObject Recognitionen
dc.subjectComputer Visionen
dc.subjectLow-shot Learningen
dc.titleLow-shot Visual Recognitionen
dc.typeThesisen
thesis.degree.disciplineComputer Engineeringen
thesis.degree.grantorVirginia Polytechnic Institute and State Universityen
thesis.degree.levelmastersen
thesis.degree.nameMaster of Scienceen

Files

Original bundle
Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
Pemula_L_T_2016.pdf
Size:
674.82 KB
Format:
Adobe Portable Document Format

Collections